Continuous progress has been made in defining a reference operating regime for the plasma, both to obtain high-performance discharges and long-duration discharges. This has made it possible to design ITER. In order to prepare for the optimal operation of ITER and its extrapolation to DEMO, the scientific community is studying the physical processes at work, and in particular the instabilities linked to plasma turbulence.
The energy supplied to the plasma to heat it to thermonuclear temperatures causes two types of instability: macroscopic instabilities that can lead to the loss of the plasma, and microscopic instabilities that generate turbulence that degrades the plasma’s insulation and therefore its confinement time.
-> Macroscopic instabilities can generate deformations on the scale of the plasma itself, causing overall displacements of the plasma towards the wall, potentially leading to cooling of the plasma or even its disappearance. These major losses of confinement (disruptions) compromise the maintenance of the reaction. Today, thanks to advances in real-time control of the tokamak configuration and plasma performance, these macro-instabilities are well under control. However, they can reappear when the plasma is pushed to its limits. In a fusion reactor for industrial use, the machine’s utilisation rate will have to be very high, of the order of 80%. Plasma disruptions and other unforeseen events must therefore be avoided. The main task will be to choose operating parameters for the installation that are sufficiently far from the limit values likely to lead to such malfunctions. In situations where such disruptions nevertheless occur, the tokamak will be equipped with systems that detect them and then limit their effects (“soft landing”). These systems have been developed on current tokamaks, and ITER will benefit from them in their entirety.
-> Microscopic instabilities generate turbulence that degrades the thermal insulation qualities of the plasma, and therefore its performance. In the plasma ring confined within the tokamak, only the core of the plasma, around 10-20% of its total volume, is molten. Temperatures here reach several hundred million degrees. The remaining 80-90% of the plasma acts as insulation between the molten part of the plasma and the walls of the torus that house it. The turbulence that increases the transfer of heat from the centre of the plasma to the edges is therefore a phenomenon that we want to minimise, in order to increase the overall efficiency of the process. The standard operating scenario for ITER is based on a regime in which turbulence is greatly reduced at the periphery of the plasma, causing what is known as a transport barrier. Increasing the effect of this transport barrier, or creating one at other points in the plasma, would increase ITER’s performance to gains (Q) of more than 10, or even to ignition.
Turbulence phenomena are still extremely difficult to model: it is in fact a multi-scale problem, both spatial and temporal, taking place in a highly complex environment: high temperatures, magnetised environment, high current densities, etc. In recent years, the increased performance of supercomputers has enabled considerable progress to be made in understanding this problem. With the foreseeable arrival of Exaflops-level supercomputers, it will soon be possible to take into account all the scales governing turbulence and thus understand its intimate mechanisms. Understanding and controlling plasma turbulence requires not only a deeper understanding of plasma physics, but also the development of diagnostic instruments that can be used to validate the modelling experimentally. This is what is currently happening on many of the world’s tokamaks.