Pedestal particle balance studies in JET-ILW H-mode plasmas

Abstract

JET-ILW type I ELMy H-modes at 2.5MA/2.8T with constant NBI heating (23 MW) and gas fuelling rate were performed, utilising ELM pacing by vertical kicks and plasma shaping (triangularity, δ) as tools to disentangle the effects of ELMs, inter-ELM transport and edge stability on the pedestal particle balance. In agreement with previous studies, the pedestal confinement improves with increasing δ, mostly due to a significant increase in pedestal density while the ELM frequency (fELM) is decreased. Improved pedestal confinement with increasing δ was observed even when the pedestal MHD stability was degraded artificially by vertical kicks, implying that increased triangularity may favourably affect the inter-ELM pedestal recovery. The workflow developed to quantify the pedestal particle balance uses high time-resolution profile reflectometry to characterise the inter-ELM evolution of the plasma particle content (dN/dt), the NEO drift-kinetic solver to evaluate the neoclassical fluxes and interpretative EDGE2D-EIRENE simulations to estimate the edge particle source. The edge particle source is then constrained by deuterium Balmer-α line intensity measurements in the main chamber, which are, however, strongly affected by reflections from the metal walls. The reflections are accounted for by the CHERAB code taking the divertor emission (the brightest light source in the torus) distribution from imaging spectroscopy measurements as input. Our analysis shows that in the second half of the ELM cycle, the volume-integrated particle source is larger than dN/dt, indicating that transport plays a key role in the inter-ELM pedestal recovery.

Effect of the isotope mass on pedestal structure, transport and stability in D, D/T and T plasmas at similar β N and gas rate in JET-ILW type I ELMy H-modes

“…The total ELM particle losses during an ELM could be estimated by volume integrating the density profiles before and soon after an ELM. However, as discussed in section 7.2 and in [65], this value is likely affected by the strong recycling of the post-ELM phase, making this estimate very uncertain. For this reason we have used a simpler qualitative approach and the total particle losses are estimated using the density drop due to the ELM, ∆n ELM e , determined from the HRTS diagnostic and discussed in figure 15(b).…”

“…Unfortunately, a reasonable quantification of the particle source using the emitted D α radiation by the main chamber plasma is not possible due to the fact that the signal is dominated by the light emitted from the divertor region and reflected by the metallic walls. A recent attempt to remove the reflected light is described in [65] showing the estimates of the particle source are characterized by very high uncertainties. As shown in [65], due to these high uncertainties no univocal quantification of the inter-ELM particle diffusivity was possible even by constraining EDGE2D-EIRENE simulations with the experimental results.…”

“…A recent attempt to remove the reflected light is described in [65] showing the estimates of the particle source are characterized by very high uncertainties. As shown in [65], due to these high uncertainties no univocal quantification of the inter-ELM particle diffusivity was possible even by constraining EDGE2D-EIRENE simulations with the experimental results. Moreover, due to the high f ELM , the inter-ELM phase of the low A eff pulses are short and hence dominated by a post-ELM fast recovery.…”

“…Moreover, due to the high f ELM , the inter-ELM phase of the low A eff pulses are short and hence dominated by a post-ELM fast recovery. Experimental evidence suggests that this fast recovery may be driven by an increased recycling flux as a result of the increased particle flux arriving to the divertor targets and limiters during the ELM crash [65,66].…”

Isotope mass scaling and transport comparison between JET Deuterium and Tritium L-mode plasmas

“…In this study, the Monte Carlo code EIRENE is used to infer the edge neutral densities based on plasma solutions calculated by EDGE2D-EIRENE for the same JET discharges 96092 (D) and 100143 (T) as analysed in the earlier sections following the JET workflow used in [46]. Cross-field drifts are not included in the simulations due to the numerical challenges that are expected when drift terms are activated.…”