SNIFfing for Negative Ions

By Ingrid Turner 

CCFE is home to two leading nuclear fusion tokamaks, JET and MAST. In order to achieve hot enough conditions required for fusion energy production and the more interesting experiments, it is vital that we have additional heating systems. One of the most important of these is the neutral beam heating system.

On JET this system has the potential to produce 35MW power in a single pulse. There are 16 positive ion neutral injectors (PINIs) on JET, and 2 on MAST. Each PINI is injected with gas which is then heated by hot filaments (metal wires). A plasma is created by putting a large electric current (arc) through this hot gas. Metal grids which are kept at a high voltage then cause the positive ions in the gas to accelerate away from the plasma. Positive ions cannot be injected straight into the tokamak, due to the large magnetic fields used. These ions are therefore neutralised first (given back an electron by sending them through more gas) and hence we have our neutral beam which is injected into the tokamak.

Unfortunately this process wastes a lot of power because only 25-30% of the ions can be neutralised, and the remaining ones have to be dumped in the form of waste heat. This gets worse for higher voltage beams, therefore it would not be economical to use these injectors on future machines such as DEMO.

Thankfully there is a solution: negative ion neutral injectors. The development of negative ion beams is crucial for fusion energy, as these have a much higher neutralisation efficiency, at around 60%. ITER, the next experimental tokamak which will be even larger than JET, already plans to use these, as it will require much higher power beams of the order 1MV as opposed to the current 125kV on JET. However there are a number of issues with negative ions, for example they have a short lifetime because their electron can be easily removed. There is also a problem with using caesium for making it easier for neutral atoms to pick up an electron, as it is a highly reactive material which is not ideal for maintenance and development of the injectors.

The Small Negative Ion Facility (SNIF) based at CCFE is a project sponsored by CCFE’s Technology Programme, and is used to study and develop negative ion beam production for beam systems on future machines beyond ITER, under work contracts for Eurofusion.  It is different to the PINIs in that it uses a radio frequency (RF) plasma source as opposed to an arc source. This works by using a flat spiral antenna which is used to excite and heat the source gas. SNIF uses high voltage grids much in the same way as they are used on PINIs, with an additional biased plate used to suppress any electrons which may get pulled through from the plasma. The final negative ion beam is then extracted and travels through a vacuum tank where it lands on a copper beam dump (mimicking a future tokamak).

There are several diagnostic systems in place on SNIF which we can use to do experiments and look at the beam power and divergence. There is a Langmuir Probe which is a measuring device inserted into the plasma and measures the plasma density, electron temperature and electric potential in the source, as well as a visible light spectrometer, which is used to measure the intensities of emission spectral lines of the hydrogen in the source. For looking at the beam, there are two cameras mounted on the vacuum tank, and finally we have the beam dump at the end of the tank which has thermocouples (temperature sensors) wired in so that we can measure the temperature rises and check that the beam is hitting the dump in the middle.

Recently I have been involved in analysing the diagnostic data, in particular from the thermocouples and the cameras. The aim of this is to obtain an accurate beam profile and to get a good idea of what the beam current is. Currently the temperature rise model we have does not fit the thermocouple model well, so I am looking into alternative materials for the beam dump. I have also conducted studies to look at how the beam width varies with different parameters, for example RF power and extraction voltage. From this we have been able to obtain the optimum parameters i.e. at which the beam width is narrowest. This has been done both experimentally and also by modelling the system and comparing the results.

In the near future SNIF will be used for many experiments, for example, testing for alternative materials to caesium in injectors, and possibly looking into energy recovery to increase the efficiency of beam systems. There are also many studies still to be conducted for the RF source which I hope to be involved in, including using the Langmuir probe to collect data. We may even try and put a tile in the way of the beam with an infrared camera as an alternative to the copper beam dump. In short, there is a huge potential for development of SNIF, and we are hopeful that ideas and results from this system will be implemented in future beam systems.