Speaker
Description
The complex stellarator magnetic field leaves a lot of room for optimisation. One of the optimisation criteria is a good confinement of the fusion-born 3.5 MeV alpha particles. Firstly, they must deposit the majority of their energy into the plasma so that the plasma stays sufficiently hot for the fusion burn to continue. Secondly, the alpha particles should not produce intolerable heat fluxes on the device inner wall.
In this work, the slowing-down process and wall loads of fusion-born alpha particles have been simulated in the GIGA stellarator using the orbit-following Monte Carlo code ASCOT5. GIGA is a stellarator currently under development by Gauss Fusion GmbH. It is a four-field-period device with 3 GWs of fusion power and a major radius of approximately 20 meters. In the studied slowing-down scenarios, it was found that the lost power fraction varied between 12 % and 23 % with higher plasma densities corresponding to lower losses. The results were obtained using four million guiding centre markers. The effect of a radial electric field on the confinement was studied in the full power scenario with the expectance that the electric field should contribute to the confinement by reducing the average radial drifts. Accordingly, it was found that neglecting the electric field increased the power losses from 16 % to 20 %.
Following the slowing-down simulations, the wall loads were investigated in the full power scenario. This was done by first following the full gyro-orbit of eight million markers until the last closed flux surface (LCFS). At the LCFS, the number of markers was multiplied so that there were 120 million markers that were then followed until a wall collision. This latter part of the simulation was carried out using the new GPU porting of ASCOT5. Largest wall power loads were observed near the toroidal angle of 45 degrees on the outboard divertors and on the wall between these divertors. A portion of the pre-conceptual divertor reached loads above 20 MW/m$^2$, with a peak load of 66 MW/m$^2$. These findings have already influenced the ongoing optimisation of the GIGA stellarator design.