Lithium surface-response modelling for the NSTX liquid lithium divertor

Abstract

We analyse lithium sputtering, evaporation and transport for the National Spherical Torus Experiment (NSTX) liquid lithium divertor (LLD) for planned high heating power plasma conditions. A temperature-dependent, data-calibrated, surface-response model is used to specify sputter yield, velocity distributions, sputtered Li ion fraction and related phenomena, for the static liquid lithium surface with D, Li and trace C impingement. Using the surface-response model, and supplied UEDGE code edge plasma parameters and LLD surface temperature profile for a typical 2 MW heating, low D-recycle shot, the REDEP/WBC code package computes Li erosion/redeposition. The results are encouraging showing negligible Li evaporation, moderate sputter erosion, and acceptable Li transport to the edge (∼7% Li/D density) and core plasma (∼1% Li/D contamination potential) and LLD adjacent surfaces. A 2% carbon plasma content does not significantly affect the LLD surface response. For fixed plasma conditions, but with hypothetically higher surface temperatures than the reference 281 °C peak, an increase to ∼350 °C appears acceptable, thus implying a significant operating temperature margin.

Lithium sputtering from lithium-coated plasma facing components in the NSTX divertor

“…Since the measured gross Y Li would lead to the complete erosion of pre-applied coatings within a few 100s ms, this is suggestive of the importance of re-deposition effects, but the relative role of accumulated vs "fresh" lithium coatings on the lithium influxes still needs to be characterized. A large prompt re-deposition fraction (resulting from the ionization mean free path being much shorter than the ion gyro-radius) was predicted by kinetic simulations with WBC-REDEP [20], and is consistent with the measurements of singly ionized lithium influxes, which are typically ∼ 1 or 2 orders of magnitude lower than the neutral lithium influxes at the OSP. Integrating the OSP neutral lithium influxes, gross lithium influxes in NSTX H-modes are estimated at ∼ 1 − 5 × 10 21 atoms/s.…”

“…Measurements of the lithium sputtering yield over a range of T sur f between 50 • C and 450 • C were obtained, with Y Li up to 0.25. In Figure 3, neutral lithium sputtering yield from a VFTRIM-3D semi-empirical model by J.P. Allain based on IIAX results [20] for deuterium incident at 45 • and 50 eV on deuterium-saturated lithium is also overlaid, showing reasonable agreement with the NSTX data (albeit with slightly higher values).…”

Measurements of core lithium concentration in a Li-conditioned tokamak with carbon walls

“…The physical mechanisms behind the different behaviour of carbon and lithium are still under investigation. A plausible hypothesis is that lithium influx from the PFCs to the main plasma is inhibited by its low ionization potential (≈5.4 eV) compared with carbon (≈11.3 eV), which favours redeposition resulting in increased retention of lithium in the divertor region, hence a reduced core contamination [27]. Other mechanisms, such as the outward diffusion of lithium caused by collisions on the higher-Z, higher density carbon ions, may also prevent lithium accumulation in the core.…”

Overview of the physics and engineering design of NSTX upgrade

“…A liquid lithium divertor (LLD) [130][131][132][133] has been tested in NSTX with a goal of assessing the ability of thicker layers of liquid lithium to extend the deuterium pumping duration relative to thin layers of solid lithium, and conceptual designs for divertor cryo-pumping systems will also be pursued for NSTX Upgrade. Such divertor particle pumping systems are not presently included in the scope of the NSTX Upgrade Project and would therefore likely be implemented following completion and initial usage of the new CS and second NBI.…”