Deuterium inventory in Tore Supra: reconciling particle balance and post-mortem analysis

Fuel retention, a crucial issue for next step devices, is assessed in present-day tokamaks using two methods: particle balance performed during shots and post-mortem analysis carried out during shutdowns between experimental campaigns. Post-mortem analysis generally gives lower estimates of fuel retention than integrated particle balance. In order to understand the discrepancy between these two methods, a dedicated experimental campaign has been performed in Tore Supra to load the vessel walls with deuterium (D) and monitor the trapped D inventory through particle balance. The campaign was followed by an extensive post-mortem analysis phase of the Tore Supra limiter. This paper presents the status of the analysis phase, including the assessment of the D content in the castellated tile structure of the limiter. Indeed, using combined surface analysis techniques, it was possible to derive the relative contributions of different zones of interest on the limiter (erosion, thick deposits, thin deposits), showing that the post-mortem inventory is mainly due to codeposition (90% of the total), in particular due to gap deposits. However, deuterium was also evidenced deep into the material in erosion zones (10% of the total). At the present stage of the analysis, 50% of the inventory deduced from particle balance has been found through post-mortem analysis, a significant progress with respect to previous studies (factor 8–10 discrepancy). This shows that post-mortem analysis can be consistent with particle balance provided specific procedures are implemented (dedicated campaign followed by extensive post-mortem analysis). Both techniques are needed for a reliable assessment of fuel retention in tokamaks, giving complementary information on how much and where fuel is retained in the vessel walls.