Validating reduced models for detachment onset and reattachment timescales
Reduced plasma models play a pivotal role in swiftly interpreting experiments, exploring different reactor design points, and providing consistency checks for high-fidelity simulations. This presentation focuses on reduced models for plasma detachment onset and transient reattachment timescales, assessing their validity across ASDEX Upgrade, JET, and MAST-U, and discussing their projection to future devices.
Experiments on ASDEX Upgrade (AUG) [1,2] and JET [3] revealed a direct correlation between the detachment threshold and the power directed to the divertors (normalised by major radius), scrape-off layer power decay width, divertor neutral pressure, and the linear summation of divertor impurity concentrations multiplied by their respective radiation efficiencies. Scenarios with an X-point radiator (XPR) [4] on AUG and JET, repeated with an abrupt increase in neutral beam power and an impurity gas cut, enabled the measurement of divertor reattachment timescales [2,3]. Observations on AUG showed reattachment timescales of ~250 ms in strongly detached XPR scenarios, significantly longer than the power modulation timescale, and ~100 ms in partially detached scenarios. The difference was attributed to the finite time required to ionise the neutrals in front of the target, measured as ~100 ms on AUG and ~1 s on JET. A reduced model predicting the ionisation timescales captured the difference between AUG and JET mainly through the major radius scaling inferred by the model.
New experimental results from MAST-U, involving NBI heated H-mode plasmas with N2 seeding and rapid oscillations between double and lower single null configurations, will be presented. These results aim to further validate the reduced models in a divertor geometry with tightly baffled conventional and extended (super-X) outer divertor legs. Achieving partial detachment of the conventional divertor through N2 seeding is demonstrated. Oscillations between double and lower single null, starting at 40 Hz, reveal that the Fulcher emission front typically returns to the target in tandem with the power modulation. This presentation explores the existence of a threshold power modulation frequency where divertor dynamics deviate from the power modulation and investigates how this deviation aligns with the reduced model’s prediction of ionisation timescales.
[1] A Kallenbach et al 2016 Plasma Phys. Control. Fusion 58 045013 [2] S.S. Henderson et al 2023 Nucl. Fusion 63 086024 [3] S.S. Henderson et al 2023 IAEA FEC [4] M. Bernert et al 2021 Nucl. Fusion 61 024001