Integrated simulations of H-mode operation in ITER including core fuelling, divertor detachment and ELM control

Abstract

ELM mitigation to avoid melting of the tungsten (W) divertor is one of the main factors affecting plasma fuelling and detachment control at full current for high Q operation in ITER. Here we derive the ITER operational space, where ELM mitigation to avoid melting of the W divertor monoblocks top surface is not required and appropriate control of W sources and radiation in the main plasma can be ensured through ELM control by pellet pacing. We apply the experimental scaling that relates the maximum ELM energy density deposited at the divertor with the pedestal parameters and this eliminates the uncertainty related with the ELM wetted area for energy deposition at the divertor and enables the definition of the ITER operating space through global plasma parameters. Our evaluation is thus based on this empirical scaling for ELM power loads together with the scaling for the pedestal pressure limit based on predictions from stability codes. In particular, our analysis has revealed that for the pedestal pressure predicted by the EPED1+SOLPS scaling, ELM mitigation to avoid melting of the W divertor monoblocks top surface may not be required for 2.65T H-modes with normalized pedestal densities (to the Greenwald limit) larger than 0.5 to a level of current of 6.5-7.5 MA, which depends on assumptions on the divertor power flux during ELMs and between ELMs that expand the range of experimental uncertainties. The pellet and gas fuelling requirements compatible with control of plasma detachment, core plasma tungsten accumulation and H-mode operation (including post-ELM W transient radiation) have been assessed by 1.5D transport simulations for a range of assumptions regarding W redeposition at the divertor including the most conservative assumption of zero prompt re-deposition. With such conservative assumptions, the post-ELM W transient radiation imposes a very stringent limit on ELM energy losses and the associated minimum required ELM frequency. Depending on W transport assumptions during the ELM, a maximum ELM frequency is also identified above which core tungsten accumulation takes place.

Kinetic modeling of ELM-induced tungsten transport in a tokamak plasma

“…Finally, in Figure 11 we look at a distribution where W is strongly screened, leading to W profiles which are hollow in the pedestal region. In this case, we obtain large particle losses to the divertor and wall, and inward penetration of 10-20% of W. From our results, it becomes clear that W expulsion by ELMs, when there is good W screening in the pedestal, will be very ineffective as also identified with a diffusive ELM model 7,31 , and a few 10 % of the W outside the pedestal can actually be transported inside it in some cases by the ELM crashes.…”

Multiple-isotope pellet cycles captured by turbulent transport modelling in the JET tokamak

“…Active research on pellet fuelling focuses on its compatibility with integrated plasma scenario constraints, including control of MagnetoHydroDynamic (MHD) modes such as Edge Localised Modes (ELMs), plasma exhaust, core turbulent transport, and desired isotope composition. Previous integrated tokamak plasma simulation (integrated modelling) including pellets focused on various aspects of the pellet cycle: improved confinement regimes [2,3,4], edge and fuelling requirements [5], the impact of fuelling on divertor heat-loads [6,7] and the extrapolation of pellet penetration and transport [8].…”

Control of the hydrogen:deuterium isotope mixture using pellets in JET

“…Finally, the departure from a 50:50% isotope mix can arise from different particle exhaust and wall recycling for deuterium and tritium and impacting the core boundary conditions. Due to all these uncertainties ITER is planning full isotope separation and two sets of pellet injectors, one for pure tritium and one for pure deuterium [6,7]. In DEMO a compromised solution with partial separation (bypass) is considered [8].…”