Dynamics and control of burning plasmas

Plasma discharges with high power amplification in a quasi-stationary state for hundred or thousand seconds are the main goal of the fusion machines of the next generation. In the research center Cadarache socalled advanced scenarios have been proposed. The confine-ment is improved by an internal transport barrier due to reversed shear (RS) produced a combina-tion of off-axis lower hybrid wave (LH) heating and central fast wave (FW) heating. The inductive current is replaced by current drive and bootstrap current.. In the steady state the plasma current is completely non-inductive. The simulations have been performed by means of the 1 1/2 dimensional transport code ASTRA. The Steering Comittee of the ÖAW-EURATOM has approved a collaboration of Austrian Re-search Center Seibersdorf (ARCS) and CEA/Cadarache to extend the physical models of the ad-vanced scenarios. In the previous investigations a prescribed normalized electron density profile has been assumed and the ion density follows from quasi-neutrality. The amplitude factor of the normalized profile is calculated by the density control system leading the plasma to a nominal fu-sion power. In reality the density is controlled by fuel injection. Thus, to obtain realistic simula-tions the transport code must be combined with a pellet injection code supplying a source for the bulk ion continuity equation which is solved by an extended ASTRA code. The actuator of the density control system is now the pellet injection frequency. The density profiles are now correctly calculated. To get realistic, results we add an improved helium model to the ASTRA code. The he-lium model consists of a two-group model for fast alphas and helium ash including recyling as an input quantity. In the previous ASTRA version just the confinement factor (effective confinemnt time over energy confinement time) has been prescribed. Furthermore we add a model for other important impurities. By this project we try to combine the flexibility of the fast running ASTRA- code with the more sophistcated models of other codes enabeling improved studies of advanced scenarios. We also attach importance to use experimentally validated transport data.

The actual project, which is accomplished within the fusion research of European Commission and which is co- financed by FWF, deals with the numerical simulation of fusion plasmas with high power amplification factor as planed for the next generation fusion machines (ITER-FEAT). The investigations are focussed on long-term scenarios, where the plasmas are heated by addi-tional heating to nominal fusion power and are maintained in that state for hundreds respectively thousands of seconds. This possibility has been offered recently by the discovery of a new plasma regime, where energy and particle confinement could be improved considerably by a particular profile of the plasma current (Internal Transport Barrier). A special control is neces-sary to maintain current profile and fusion power in a steady-state. By means of the transport barrier not only the confinement of the plasma fuel (deuterium and tritium), but also the confinement of impurities, especially of helium, which is produced by fusion reactions, is improved. The helium generated by fusion reactions is referred to as helium ash, because it is a by-product of burning. It is to be demonstrated, that the presence of an internal transport barrier does not lead to an strong accumulation of helium ash, which is inconvenient for reactor operation. In our project we have built a helium diffusion model into the internationally used plasma simulation code ASTRA. The accumulation of helium ash as well as the additional heating power necessary to reach the nominal fusion power has been calculated. The recycling factor of helium at the plasma wall represents an important parameter for the scenarios, which should be investigated closer by experiments. The same is the case for boundary conditions and open factors in the diffusion models. Nevertheless, the results show, that on plausible assumptions the internal transport barrier represents no essential obstacles to tokamak operation. The plasma fuelling can be effected by means of frozen deuterium-tritium spheres with a diame-ter of some millimeters, referred to as pellets, which are injected into the plasma at high veloci-ties. The pellets are dissolved within a few milliseconds and transformed into plasma. A pellet simulation code developed in the research center Cadarache has been combined with our plasma simulation code and used for investigations of the long-term scenarios mentioned above. It has been shown that the injection frequency can be used for control of density and fusion power. Furthermore, we investigated on which conditions the necessary penetration depths of the pellets into the plasma by means of available technology.