Stable plasma for ITER

IPP develops world's best code for stability calculations / tokamak theory aided by the stellarator

June 18, 2007
The operation of future fusion power plants requires that the hot fuel be confined with good thermal insulation. The magnetic cage used for this purpose therefore has to be as robust as possible. And it is just with the particularly interesting "advanced" modes of operation planned for the ITER international fusion test reactor that perturbing instabilities are expected: So-called external kink modescould severely reduce the fusion yield attainable with ITER. Possible remedies have now been investigated by Max Planck Institute of Plasma Physics (IPP) on behalf of the EU. The new Starwall computer program developed for this purpose can be regarded as the best of its kind in the world – credit being due to the expertise of the stellarator specialists at IPP that has gone into the making of this tokamak code.

The objective of the world-wide efforts expended on nuclear fusion is to develop a power plant which, like the sun, derives energy from fusion of atomic nuclei. To ignite the fusion fire the fuel, a hydrogen plasma, has to be heated to temperatures exceeding 100 million degrees. The next major step in this research is the ITER (Latin for "the way") international test device. It is to deliver in lasting pulses a fusion power of 500 megawatts – ten times as much as is needed to heat the plasma. Construction of the device is to commence next year at Cadarache in France.

Distracting instabilities
In order to maintain the high temperature, one must succeed in confining the fuel in magnetic fields without contacting the vessel wall and with good thermal insulation. The complex interaction between the plasma particles and magnetic cage, however, makes possible a whole series of instabilities which impair confinement. Though of no major significance in previous devices, external kink instabilities are extremely undesirable in the ITER fusion test reactor because they deform the confining magnetic field: The ensuing hose-like bumps and dents on the exterior of the plasma impair confinement, thus reducing the fusion yield. In ITER they would – according to IPP’s calculations – occur in the very plasma states that are being counted on to develop a tokamak capable of continuous operation.

There are, however, possible remedies: If confined by an infinitely conducting – i.e. superconducting – wall, the instabilities could not form in the first place. This is because the plasma motion induces electric currents in the wall whose magnetic field counteracts the cause: The plasma is stabilised. A "normal" steel wall can at least slow down the formation of kink instabilities – from microseconds to milliseconds. This makes the process slow enough for an automatic feedback system to intervene. Weak electric control currents flowing in small magnet coils attached to the wall can already "capture" and eradicate the bumbs and dents before they grow. They simulate, as it were, the magnetic response of a superconducting wall to the motion of the plasma.

The Starwall code
For this to function technically one must be able to describe and calculate the processes precisely. It is here that calculations for fusion devices of the tokamak type – to which ITER belongs – usually benefit from the simple structure of these experiments. The ring-shaped plasma vessel, the magnetic cage ring, and the ring-shaped plasma confined therein are of axisymmetric form: There are no changes taking place around the ring. Accordingly, it is therefore mostly sufficient to calculate in just two spatial dimensions. If, however, the electromagnetic interactions between the plasma and vessel wall are to be described, it has to be taken into account that notwithstanding the general symmetry the wall is not identical everywhere. In some places there are large ports for making the plasma accessible for heating facilities, pumps and measuring instruments. Exact calculation of kink instabilities and stabilisation of these by the wall thus call for inclusion of all three spatial dimensions.

This is just what the newly developed Starwall code of IPP can do. It describes – for the first time in the world – the plasma and vessel walls in all three spatial dimensions. Here tokamak theoreticians have benefited from collaboration with their stellarator counterparts: To wit, stability codes for fusion devices of the stellarator type – such as Wendelstein 7-X being built at the Greifswald branch of IPP – are always three-dimensional, for stellarators with their bizarre coil system do not have the simplifying symmetry of tokamaks. Without this preliminary work of stellarator theoreticians development would have been much too costly.

The calculations with Starwall for ITER show: With a feedback system ITER discharges could remain stable up to a plasma pressure that – depending on the pressure and current profiles in the plasma – would be 50 per cent higher than it would have been without stabilisation: a huge step forward. Experimental investigations in Garching’s ASDEX Upgrade tokamak device are planned: With 24 control coils and a wall close to the plasma it is to be investigated for ITER how readily kink instabilities can be influenced

Isabella Milch

Go to Editor View