(TH/P318) Kinetic Simulation of Heating and Collisional Transport in a 3D Tokamak
A. Bustos Molina^{1)},
F. Castejón^{1)4)},
L. A. Fernández^{2)4)},
V. MartinMayor^{2)4)},
A. Tarancón^{3)4)},
J.L. Velasco^{3)4)}
^{1)} Laboratorio Nacional de Fusion Asociacion EURATOM/CIEMAT para Fusion, Madrid, Spain
^{2)} Dep. de Fısica Teórica I. Universidad Complutense, 28040Madrid, Spain.
^{3)} Dep. de Fısica Teórica. Universidad de Zaragoza, 50009Zaragoza, Spain.
^{4)} BIFI: Instituto de Biocomputación y Fısica de Sistemas Complejos, 50009Zaragoza, Spain
Abstract. The microwave plasma heating has a strong influence on collisional transport,
experimentally observed both in stellarators and tokamaks. The estimate of the interplay
between heating and collisional transport implies solving a 5D kinetic equation (2D in
momentum space and 3D in real space), which is a difficult task. We attempt to solve this
problem using a recently developed code (ISDEP: Integrator of Stochastic Differential
Equations for Plasmas) in a tokamak with ripple as a test device. Previously, ISDEP has been
successfully applied to the TJII
stellarator [1].
Our Monte Carlo method is based on the equivalence between the linear FokkerPlanck
and Langevin equations. This allows us to describe the system by taking average values over
many independent test particle trajectories. The dynamics of these ions are determined by the
guiding center approximation, ionion
collision [2] and the interaction with microwaves (Ion
Cyclotron Heating, ICH, since we are dealing with ion transport). We also have developed a
selfconsistent
method to update the background temperature in order to introduce nonlinear
terms. Thus, we have modified ISDEP to include the geometry of a tokamak with ripple and the
Langevin equations for ICH [3].
The European Computer Grid (EGEE) has been used to perform the calculations. The
main results of this work are the calculation of transport quantities and the velocity probability
distribution function. We have compared these results in the cases with and without heating, and
investigated the differences between them. There are three main conclusions: i) the increment of
the kinetic energy, with a consequent increment of the temperature, and an increase of the
outward fluxes, which implies a reduction of the particle confinement. ii) The deviations of the
distribution function from the Maxwellian, both in the bulk and in the wings in the presence of
ICH. iii) The included ripple is not enough to generate toroidal asymmetries.
This computer code can be adapted to other geometries and allows to consider other
features that can be taken into account in the ion dynamics (NTM, ee
collisions and Alfvén instabilities).
[1] F. Castejón et al, Nucl. Fusion 42, 271 (2002); [2] Z. A. Pietrzyk et al. Physical Review Letters 86, 1530 (2001); [3] S. Murakami, Nucl. Fusion 45, 221 (2003)
Full paper available (PDF)
