Overview of fuel retention and recovery in JET DT operation
Tritium inventory build-up is a safety issue for next step devices. JET brings a unique contribution on fuel retention and recovery in a metallic device, as it has operated both in deuterium (D) and deuterium-tritium (DT) plasmas. This paper will provide an overview of results from JET in this field, with a focus on recent DT campaigns (DTE2 in 2021 and DTE3 in 2023). Global fuel retention from particle balance and new in-situ fuel retention local measurement using Laser Induced Desorption with gas detection using Quadrupole Mass Spectrometers (LID-QMS) will be reported. The results presented here are the first demonstration of LID-QMS, presently foreseen to assess the in-vessel fuel inventory in ITER, in a tokamak operating with tritium. The experimental setup to evaluate particle balance in JET during DT operations will be described (in-vessel pVT method, accounting for injected and exhausted gas to derive fuel retained in the vessel walls). The main features of the LID-QMS diagnostic, implemented on JET for the DTE3 campaign, are presented. The best conditions to operate the LID-QMS (low vessel base pressure and low torus pumping) will be detailed. The same procedure has been applied during DTE2 and DTE3 to assess global particle balance and derive fuel retention (in-vessel pVT following a series of identical L mode DT discharges to load the walls). Results from both campaigns are compared, with a focus on the main sources of uncertainties. In particular, the impact of impurities in the exhaust gas during the DTE3 experiment (4He from previous gas calibration, 3He from the T injection, Ne from previous impurity seeding experiments) is discussed. In addition, during DTE3, regular measurements with LID-QMS were performed throughout the campaign to monitor the evolution of the fuel retained in tile 1 and tile 0 of the divertor, where most of the fuel retained in the divertor is expected to be located, based on past post mortem analysis results. By regularly cleaning a dedicated area with the laser, both the fuel retained after ~1 week of DT operation and the total cumulative fuel since the beginning of the campaign could be monitored. Clear RGA signals could be obtained after laser firing on masses 3 to 6 (expected to correspond mainly to HD, D2, DT, T2 respectively), providing insight on local co-deposition of fuel during operation for the first time. Fuel retention derived from gas balance during DT operation in DTE2 and DTE3 is discussed in the light of previous gas balance measurement performed during D operation with the JET metallic walls, as well as the results from post mortem analysis. Finally, a clean-up campaign was performed after DTE2 and DTE3 to recover T trapped in the walls, using successive cleaning methods (baking at 320°C, Ion Cyclotron Wall Conditioning, cycling limiter pulses, diverted pulses with Raised Inner Strike Point to reach areas with deposited layers in the inner divertor, and finally high power pulses). During the DTE3 clean- up phase, regular particle balance and LID QMS measurements were used to monitor the T recovery. The implications of the JET results in terms of fuel retention and recovery will be discussed for ITER (including moving to a full tungsten first wall) and plans for DTE3 post mortem analysis of JET plasma facing components presented to complete the fuel retention overview.