Physics of plasmas; works, studies, applications

 Radial current and plasma rotation produced by ionization of the neutral atoms (gas-puff, pellets, impurity seeding)

This is a mechanism by which the simple ionization process has a direct effect on the confinement. The ionization creates a radial current, i.e. a torque which can be sufficiently strong to overcome the poloidal magnetic damping. The induced plasma rotation is sheared and this generates a barrier to the turbulent transport.

The torque/rotation created by the ionization events will interfere with a pre-existing rotation. Then the effect can be synergetic and the rotation (further: confinement improvement) is enhanced; alternatively, the rotation tendencies (pre-existing and from ionization) can be opposite and this is detrimental to the confinement.

It is a mechanism of neoclassical origin, easy to understand if we represent the event of ionization as an "initial value problem", instead of the usual bounce averaging.

Here is the Abstract of the arxiv.org preprint.

A change of the particle density (by gas puff, pellets or impurity seeding) during the plasma discharge in tokamak produces a radial current and implicitly a torque and rotation that can modify the state of confinement. After ionization the newly born ions will evolve toward the periodic neoclassical orbits (trapped or circulating) but the first part of their excursion, which precedes the periodicity, is an effective radial current. It is short, spatially finite and unique for each new ion, but multiplied by the rate of ionization and it can produce a substantial total radial current. The associated torque induces rotation which modifies the transport processes. We derive the magnitude of the radial current induced by ionization by three methods: the analysis of a simple physical picture, a numerical model and the neoclassical drift-kinetic treatment. The results of the three approaches are in agreement and show that the current can indeed be substantial. Many well known experimental observations can be reconsidered under this perspective. In reactor-grade plasma the confinement can be strongly influenced by adequate particle fuelling.

 Filamentation of the current/vorticity narrow edge layer in the H-mode and the Edge Localized Modes

This is a simple application of the idea of break-up and filamentation of a layer of density and current as result of a "tearing" instability, according to the theory of Bulanov and Sasorov (Sov. J. Plasma Phys. 4 (4, July-August 1978) 1979 page 418). We have proposed this as an alternative to the "peeling-balloonning" instability in the case of the Edge Localized Modes, during a meeting at JET

Short presentation at TTG Culham 2009.

A poster presentation at TTG Cordoba 2010.

The theory is connected with a subtle, apparently distant, problem, - of systems equivalent to a gas Chaplygin with anomalous polytropic. The instability is universal and leads to formation of singularities. The solution that is applicable in the case Bulanov-Sasorov has been derived by Trubnikov, Zhdanov (Phys. Rep).

Detailed explanations can be found in our work "A model for the reversal of the toroidal rotation in tokamak" (F. Spineanu and M. Vlad, Nuclear Fusion 52 (2012) 114019 ) or the preprint on arxiv.org

 Reversal of toroidal rotation

Data from ALCATOR C-MOD show that in certain condition there is reversal of the direction of the plasma toroidal rotation. There may be several sources for this phenomenon. One of them is a transient but vigurous neoclassical process that acts on the trapped particles. We start from the possible onset of rolls of convection having the same diraction of rotation. In the poloidal plane they are like a set of tangent circles. We then note that the envelope of the circumference of these circles is a closed poloidal streamline (preserving some oscillations that are due to the discrete circles that are the boundaries of convection cells). The poloidal flow is triggered on a fast time scale which is comparable with the bounce time. Then on one side of a banana the ions feel a field and on the other side they feel another, which is commonly known as the effect of neolassical polarization. A parallel flow is induced in this way. Quantitatively it looks reasonable.

This is the arXiv.org preprint. It has appeared in NF.

A poster presentation in 2011.

This scenario is inspired by a similar physical system, the Ranque-Hilsch tube. This is commercial, it is very common, but, Incredibly as may seem, it is not understood at this time, except if we (1) accept a giant viscosity, or (2) accept that there are convection rolls developing inside. Even so it is difficult.

 Zonal flow sustained by vortices coming from the connex turbulent field

We are interested in the system consisting of a region of turbulence (of drift wave) bounded on one side by a laminar sheared flow, which is a zonal flow. The turbulent region produces vortices of different amplitudes. The neighbor zonal flow creates a gradient of vorticity and the vortices move spontaneously on this gradient of vorticity: those that have a particular sign are attracted to the maximum of the vorticity and the opposite ones are repelled. A vortex comes with an internal content of rotation. The "melting" inside the sheared laminar flow sustain the shear. This interesting fact is known in fluid physics but one of the best analysis comes from plasma (Schecter and Dubin). It is a kind of feeding a flow by pieces of vorticity which become sources of momentum. The continuum version of this is Reynolds stress.

This is the arXiv.org preprint.

A poster presentation.

This work is not entirely speculative (as it may appear).

The initial intention was to discuss the neoclassical effect on the banana, consisting of toroidal acceleration due to fast transient poloidal flows. A random collection of events consisting of generation of sets of convection rolls would ensure a kind of statistical stationarity of the toroidal rotation. The problem is however more general.