Kein Folientitel

1
Modelling with the 2D Multifluid Code TECXY of
TEXTOR Discharges in the Presence of DED
H.Gerhauser (1) , G.Telesca (2), R.Zagórski(3)
(1) Institut für Plasmaphysik, Forschungszentrum
Jülich GmbH, EURATOM Association,
Trilateral Euregio Cluster, D-52425 Jülich,
Germany (2) Department of Applied Physics, Ghent
University, Rozier 44, B-9000 Gent, Belgium (3)
Institute of Plasma Physics and Laser
Microfusion, EURATOM Association, P.O. Box
49, Hery Street 23, 00-908 Warsaw, Poland
OUTLINE

• Motivation
• Physical model  
• ? Multifluid description  
• ? Stochastic transport
• Experimental setup and observations
• Simulation of effects of shifting the plasma from
  ALT-II to DED
• Modelling of plasma parameters with DED target
  plates
• in the shadow of ALT-II limiter
• 6. Summary

2
MOTIVATION

• TECXY successfully used to model TEXTOR with
  ALT-II and bumper limiter
• DED configuration geometrically similar to the
  old bumper configuration
• Currents in DED coils produce perturbation layer
  with additional stochastic transport
• Two modes of operation up to now
• ? 3/1 mode with large islands far in the core,
  which can not be simulated by 2D boundary code
• ? 12/4 mode characterised by very short
  penetration depth, such that the plasma column
  has to be shifted horizontally towards the HFS.
  This implies a change of the location of
  recycling and of impurity source (main plasma
  sink at the ALT-II limiter is replaced by the
  sink at the DED target plates)
• With TECXY we can separate the effects of plasma
  shift from switching on the stochastic transport
• Need for interpretation of experimental data
• Possibility to model simultaneous action of DED
  and ALT-II (like in future 6/2 mode ?)

3
BASIC MODEL ASSUMPTIONS

• 2D MULTIFLUID DESCRIPTION OF THE PLASMA -
  Braginskij-like equations
• CLASSICAL TRANSPORT ALONG FIELD LINES (21 -
  moment Grad approximation)
• RADIAL TRANSPORT
• anomalous constant diffusion coefficients,
  Alcator-like scaling
• TWO TEMPERATURE MODEL
• all ions have common temperature
• EQUATIONS FOR DRIFTS AND CURRENTS
• ELECTRIC FIELD FROM OHMs AND KIRCHHOFFs LAW
•  Global ambipolarity of radial currents in the
  transition layer
• ATOMIC PROCESSES ionization, recombination,
• excitation, charge exchange (Carbon impurity)
•  NEUTRALS (Analytical model for neutrals)

TEXTOR Tokamak Geometry
4
2D Model Equations
5
INTEGRATION DOMAIN AND BOUNDARY CONDITIONS
DED
Sheath conditions with drifts
6
TRANSPORT IN STOCHASTIC FIELD

• The transport in stochastic field is considered
  as a sequence of displacements both parallel and
  perpendicular to the magnetic field lines, which
  are linked together forming "optimal paths" with
  largest effective radial transport (Tokars
  model)
• Essential parameters are DFL (field line
  diffusivity) and LK (Kolmogorov length)
• The radial transport coefficients due to
  anomalous transport have been modified to account
  for the contribution from stochastic transport

• Additional impurity diffusion from stochastisity
  is very small
• Friction with the background plasma flow leads to
  significant increase
• of radial impurity convection

7
TRANSPORT IN STOCHASTIC FIELD

• Significant increase of the radial heat
  conductivity due to stochasticity (index DED)
• (related to the large classical parallel heat
  transport)

• Strong increase of radial particle convection
  from stochasticity in comparison to the classical
  radial drift velocities

Classical radial drift velocities of the carbon
ions are usually negative thus favouring
transport away from the DED surface. Radial
convection due to stochasticity is directed
towards DED and can be significant, in particular
for higher Z ions.
DFL 2e-5 m, LK10 m
8
EXPERIMENTAL SETUP
Ratio of the brilliances of C2 and C4 lines
along the different chords for a horizontally
shifted plasma with respect to standard position
O Almost no change for C4 O The
location of radiation from . C2 changes
significantly with . horizontal shift, while
the total . C2 radiation remains essen- .
tially unchanged

• Brilliance of C2 and C4 lines in the UV
  measured simultaneously along chords 1-9 at
  repetition time of 100 msec
• C2 line measured along A and B
• Average Zeff from line-integrated Bremsstrahlung
  in the visible
• Total radiated power Prad by bolometry (Y.Liang)

9
EXPERIMENTAL RESULTS
0.35 MW lt Pinp lt 1.1 MW 2.8 lt qedge lt 3.6 2 cm lt
DR lt 4cm 7kAlt IDED lt 12kA

• Each of C2 points is the ratio of the averaged
  value of C2 lines on chords 1-5. For the C4 all
  chords are used
• Radiation from C2 ions increases by the factor
  2-3 on the HFS when the plasma column is
  horizontally shifted. It increases only up to a
  factor 1.5 when the DED field is added to the
  shift
• C4, Zeff and Prad remain nearly constant with
  shifting plasma and with adding the DED
  stochastic field.

10
SIMULATION RESULTS

• Integration zone covers region corresponding to
  the 9 lines of sight of the multi-chord
  diagnostic
• For two density levels we calculate averages of
  Zeff and of line radiation
• Shifting from ALT-II to DED increases hydrogen
  and lower Z carbon ion radiation but slightly
  reduces Zeff and higher Z radiation (in the edge)
• Stochasticity affects only higher Z ions (?)

11
PLASMA SHIFT AND STOCH. TRANSPORT
SHIFTED DED
NORMAL ALT-II
SHIFTED

• The effect of shifting much more pronounced than
  effect of stochastic transport
• Background plasma densities and temperatures far
  from target very similar
• Close to the target strong increase of recycling
  and peaking of density

12
PLASMA SHIFT AND STOCH. TRANSPORT
NORMAL ALT-II
SHIFTED
SHIFTED DED
WALL
DED-down
WALL
WALL
DED-down
e-side
DED-up
DED-up
i-side
CORE
CORE
CORE

• Position of the Zeff maximum (high Z ions) almost
  the same (90o-135o)
• No essential changes due to stochastic transport
  on plasma and impurities
• Slight screening of high Z impurities due to
  stochasticity

Radially averaged
Tendencies and conclusions are largely
independent on the density regime and anomalous
transport coefficients
13
PLASMA SHIFT AND STOCH. TRANSPORT

• Position and distribution of low Z ions
  determined by the position of the sink
• Position of high Z ions defined by interplay
  between flows and thermal forces
• Change to the poloidal flow (GCL disappears)
• Change to plasma pressure and parallel flow
• Stronger friction shifts high Z ions from
  stagnation point towards upper side of DED

CORE
WALL
e-side
WALL
DED-down
i-side
DED-up
CORE
CORE
14
DED Target plates in the shadow of ALT-II limiter
IDED off
IDED on

• Main sink at ALTII-limiter
• DED target plates in the shadow, close to
  separatrix (D0.5 cm)
• High density regime

ALTIIe-side
ALTIIe-side
CORE
CORE
DED
DED
High recycling regime develops
WALL
WALL
ALTII i-side
ALTII i-side
CORE
CORE
ALTII i-side
ALTII i-side

• Strong effect of switching on IDED
• Increase of Ne and decrease
• of Te
• Recycling zone extends from DED up to ALTII
  i-side (involves neutrals and global circulation)

ALTIIe-side
ALTIIe-side
WALL
WALL
15
DED Target plates in the shadow of ALT-II limiter
Stochastic transport on
Broad local maximum of Ne and minimum of Te
16
DED Target plates in the shadow of ALT-II limiter

• Density of C4 and C5 reduced
• Density of low Z ions (C2, C3) increased, but
  not necessarily radiation in the edge
• Strong reduction of Zeff

Screening efficiency is improved mainly due to
the higher plasma density indirect effect of
stochastic transport

• This favourable high recycling zone
• develops only if
• the distance between DED target plates and
  separatrix is small
• density is high enough

17
DED Target plates in the shadow of ALT-II limiter
IDED on
IDED off

• Main sink at ALTII-limiter
• DED target plates far in the shadow (D2 cm)
• High density regime

• Relatively small effect of
• stochasticity
• reduction of Ne and Te
• improved screening

MARFE-like structure (similar to bumper limiter)
18
DED Target plates in the shadow of ALT-II limiter
IDED off
IDED on
ALTIIe-side
ALTIIe-side
CORE
CORE

• Main sink at ALTII-limiter
• D0.5 cm
• Low density regime

DED
DED
WALL
WALL
ALTII i-side
ALTII i-side
CORE
CORE

• Small effect of stochasticity on plasma and
  impurities
• Distance between DED target plates and separatrix
  not important

ALTII i-side
ALTII i-side
ALTIIe-side
ALTIIe-side
WALL
WALL
19
SUMMARY

• With the 2D-code TECXY we simulated some basic
  features of TEXTOR-DED (static fields)
• Stochasticity in the DED region was described by
  a model for optimal paths and increased locally
  radial transport
• Shift of plasma from ALT-II to DED has stronger
  effect on discharge than switching on stochastic
  transport
• Comparison of C2 and C4 line radiation
  intensities with experimental observations shows
  similar tendencies
• With DED slightly in the shadow of ALT-II and at
  sufficiently high densities a favourable
  high-recycling regime develops with good
  screening of impurities