2 frequency gyrotron manufactured by GYCOM, Russia, used for ASDEX Upgrade. Photo: IPP, Garching

Aiming to develop solutions that can lead to a more flexible Electron Cyclotron Resonance Heating (ECRH) for ITER, an international network named “Advanced ECRH for ITER” has been formed. The “virtual institute” is funded with 900.000 Euros for 3 years by the German Helmholtz Society and is headed by the Max-Planck-Institut für Plasmaphysik (IPP) in Garching, Germany. Network partners are the Forschungszentrum Karlsruhe (FZK), the Universities of Stuttgart and Karlsruhe, as well as the Institute of Applied Physics of the Russian Academy of Science in Nizhny Novgorod and the Instituto di Fisica del Plasma in Milano.

The network has two main goals, one of them is to improve the frequency range of the gyrotron heating system. Within ITER, a single frequency gyrotron operating at 170 Gigahertz is foreseen. “But a single frequency gyrotron reduces the operational margins, especially regarding the toroidal magnetic field”, explains Prof. Hartmut Zohm, speaker of the new virtual institute. “If we had the opportunity to operate the gyrotron at multiple frequencies, we could reach any region of the plasma and for example fight the Neoclassical Tearing Modes (NTMs) much more effectively. “NTMs are instabilities driven by the pressure of the hot plasma in a fusion machine. As such they are a major concern for ITER and any tokamak as they spoil the confinement and cause disruptions.”

But how to tell a gyrotron to work at various frequencies? “ITER has decided not to face this challenge at present as a more flexible source means more complications for the project as a whole”, Zohm says. “So this option was not considered in the ITER Design. Though we all agree that it would be nice to have such a tool heating ITER.” For an independent network like the virtual institute, the task is easier to solve. In a first step, a two-frequency gyrotron was implemented in the ASDEX Upgrade tokamak, working at 105 and 140 Gigahertz. The experiments delivered a maximum pulse of close to 1 Megawatt and 10 seconds duration. Within the course of 2007, IPP expects to implement a new four-frequency gyrotron operating at the intermediate frequencies of 115 GHz and 127 GHz as well.

So what would be the benefit of these experiments for ITER? How would a multiple- frequency gyrotron have to look like to fit into the ITER machine?

“Talking about the physics, we know that it would help a lot to control the NTMs, but the profit of course has to be quantified in experiments first. Regarding design issues, the ITER launchers or antennas should be compatible for different frequencies, the same holds for the transmission lines. Whether the optical mirrors can cope with multiple frequencies still has to be proven.”

When would it be possible to implement this new system on ITER? “As the lifetime of a gyrotron usually is around 5 to 6 years, a new multi-frequency gyrotron could be implemented thereafter, in the second round of the experiment.”

The second aim of the network is to improve the modulation scheme for NTM stabilisation. From experiments on ASDEX Ugrade, it looks like injecting phased ECCD (Electron-Cyclotron Current Drive) into the NTM islands will have a clear benefit in terms of power requirements for stabilisation. Since NTMs usually rotate with respect to the ECCD launching antenna, the former approach meant to switch the gyrotrons on whenever an NTM passes by and to switch it off when it has passed. But under these circumstances the gyrotrons would only operate with 50 percent duty cycle. The new approach reads as follows: If one had a fast switch to send the ECCD power to several antennae distributed around the torus, the gyrotron could operate with 100 percent efficiency. But since ITER expects mode rotation frequencies above 1 kHz, a mechanical switch is out of question. A solution to this problem could be the use of a diplexer based on interference in conjunction with a slight frequency tuning of the gyrotron. The diplexer would send the ECCD power to the wanted antenna, depending on the frequency shift. First results obtained with such a scheme at IPP Greifswald are very promising and the network will study conceptually how to implement such a system in the ITER ECRH system, potentially doubling its NTM stabilisation capability.