Speaker
Description
Power exhaust remains a critical challenge for ITER, DEMO, and next-generation tokamaks, with unmitigated exhaust fluxes in an ITER-like device estimated at 100MWm−2, for which real-time control is essential. Maintaining detachment is further complicated in reactors by transients, caused by pellet fuelling and MHD perturbations (sawteeth, LH/HL transitions, unsuppressed ELMs, ...), resulting in intense divertor power bursts. These can ’burn-through’ the detached neutral divertor buffer, reattaching it on sub-millisecond timescales, resulting in target melting [1]. Although ELM suppression is essential for reactor-scale devices, such as ITER [1], some transients are unavoidable in steady-state reactor operation; a quantitative understanding of the power losses and processes in the divertor during burn-through of the detached buffer is essential. To characterise this evolution, diagnostics with high temporal resolution for intra-transient analysis are required, paired with analysis tools for a quantitative understanding of plasma-neutral interactions [2-4].
Bayesian analysis techniques for chordally-integrated spectroscopic data was successfully used to infer hydrogenic processes (ionisation, Electron-Ion Recombination, Molecular Activated Recombination / Dissociation) and their power losses in quasi-steady-state conditions [2]. This work prepares these techniques for application to MAST-U’s UltraFast Divertor Spectrometer diagnostic (UFDS) [5], operating at 800 kHz, enabling intra-transient estimates of ionisation sources, associated power losses and recombination. This provides insight into the physics of ELM burn-through, enabling validation of full and reduced [5] models for transient burn-through. Developing reduced analysis techniques to prepare real-time line-integrated sensors for ELM burn-through is considered. However, detachment is inherently a 2D phenomena, requiring camera diagnosis for 2D analysis. With multi-diagnostic Bayesian integrated data analysis techniques, this enables inferences of 2D parameter maps in the divertor (Te, ne, ...) [3,4], using cameras. These, along with fast multi-wavelength imaging (10 kHz), are planned to obtain 2D insight into ELM burn-through.
The aim is to develop tools for quantitatively understanding transient burn-through in the detached divertor, with an emphasis on identifying dominant hydrogenic radiative loss channels and processes in the intra-transient detached neutral buffer, contributing to an improved understanding of transient interactions within a detached divertor volume to improve reactor-scale extrapolations and develop sensors that provide a pre-warning signal for transient burn-through.
[1] Paschalidis, et al. 2024 Nucl. Fusion, 64 126022
[2] Verhaegh, et al. 2024 Plasma Phys. Control. Fusion, 63 035018
[3] Greenhouse, et al. 2024 Plasma Phys. Control. Fusion, 67 035006
[4] Bowman et al 2020 Plasma Phys. Control. Fusion 62 045014
[5] Flanagan et al 2025 Nucl. Fusion 65 116031