Self-consistent modelling of a liquid metal box-type divertor with application to the divertor tokamak test facility: Li versus Sn

Self Cite

Accidental transients in fusion reactors, such as the in-vessel Loss-of-Coolant Accident (LOCA) considered here, are generally analyzed using system-level codes. However, because of their lumped nature, such tools cannot predict local pressure and temperature values, which are instead of interest to assess the integrity of the Vacuum Vessel (VV) structures. It is then fundamental to prove that the system-level tools' predictions are at least conservative. In this work, we analyze the helium flow inside the VV following a LOCA in the helium-cooled blanket of the EU DEMO, using a 3D transient CFD model implemented in the commercial STAR-CCM+ code. In view of the large pressure ratio, a hypersonic flow regime (Ma > 5) develops, with the formation of shock fronts requiring a high time and space resolution. The model is applied to compute the evolution of the pressure distribution inside the VV and then to compare it with the information provided by the system-level GETTHEM model. The predictions by the two models of the intervention time of the VV pressure suppression system are compared, showing that the 0D model is conservative.

Section: Iiia2 Initial Conditionsmentioning

confidence: 99%

3D transient CFD simulation of an in-vessel Loss-Of-Coolant Accident in the EU DEMO fusion reactor

Zappatore

Froio

Spagnuolo³

et al. 2020

Self Cite

“…Roughly, the material consumption in a shot 100 s long is r Li ≈ 0.115 kg for Li and r Sn ≈ 0.02 kg for Sn. This problem is being considered, as mentioned above [8].…”

Section: Global Parametersmentioning

confidence: 99%

“…Conversely, resilience to edge-localized modes, periodic transient strong losses of particles and energy from the core, is still being investigated. The problems linked to too strong evaporation, namely the recovery of the lost material and plasma pollution, are the objects of several ongoing studies for both I-DTT [8] and DEMO [9].…”

Section: Introductionmentioning

confidence: 99%

Perspectives for the liquid lithium and tin targets in the Italian Divertor Test Tokamak (I-DTT) divertor

Pericoli‐Ridolfini¹,

Ambrosino²,

Chmielewski³

et al. 2019

The behaviour of the scrape-off plasma of the projected Italian tokamak I-DTT is analysed by means of the two-dimensional edge code TECXY for either Li or Sn as the liquid target material. A scan in the outer midplane separatrix density range n e,omp ≈ (0.5-1.1) × 10 20 m −3 is carried out for several power inputs into the scrape-off layer (SOL) up to a maximum of P SOL = 35 MW. The radiative capability and peak load mitigation are higher for Sn than for Li, but only slightly so at high density, so that the entire density range can be explored with liquid Sn targets, unlike Li whose employment at low density and high power is questionable. However, if a ceiling to the Sn concentration is set for high core fusion performance, Sn mitigation at low density may drop below Li. Nonetheless, concerns about the Li option derive from the very large material consumption and from the fact that the main mechanism of impurity release is sputtering, which is outside the full control of the operator. Considering specifically the three main scenarios-low, medium and high density-at the maximum auxiliary power of 45 MW, operations are successful with Sn, whereas Li requires an additional impurity able to radiate in the core, except at the highest density. The snowflake plus configuration, compared with the standard single null, as a sample for advanced divertors, appears more suited to exploit the advantages of the liquid metal option. Concerning additional impurities, only Ar and Ne are considered at present. They can strongly help in reaching and studying deep detached conditions even at the maximum power, but are not essential for operations. Wide flexibility in designing the target refrigeration system is required to fully exploit its great potential.

“…For a given plasma heat and particle flux impinging on the target, a larger mass flux is eroded from a Li target with respect to a Sn target, due to the larger sputtering yield and vapor pressure. Conversely, for a given amount of eroded metal, Sn leads to larger plasma cooling, mostly associated to line radiation and bremsstrahlung [14]. For both Li and Sn, the vapor shielding effect has been experimentally observed [15,16].…”

Section: Introductionmentioning

confidence: 99%

“…Modeling the scrape-off layer (SOL) plasma in the presence of an LMD is essential to predict whether an operating window for this system exists, both in terms of core plasma compatibility and of target heat load. This challenge has been addressed in the past by means of simplified models in [14,[21][22][23]. Recently, more detailed simulations, including the transport of the eroded metal in the SOL and a detailed account of its interactions with the plasma, have been performed by means of 2D edge plasma codes.…”

Section: Introductionmentioning

confidence: 99%

SOLPS-ITER simulations of a CPS-based liquid metal divertor for the EU DEMO: Li vs Sn

Nallo¹,

Mazzitelli²,

Moscheni³

et al. 2022

Self Cite

In this work, we study the effect of installing a liquid metal divertor (LMD) using a capillary-porous structure in the EU DEMO tokamak within the same envelope of the baseline solid divertor. We used the SOLPS-ITER code to model the Scrape-Off Layer (SOL) plasma and neutrals, coupled to a target thermal model to enable the self-consistent calculation of the LM target erosion rate, and adopting a fluid neutral model for the sake of simplicity. First calculations considering only D and Li (or Sn) showed a significant reduction of the steady state target heat load with respect to simulations considering only D, thanks to vapor shielding. Nevertheless, the computed peak target heat flux (~31 MW/m2 and ~44 MW/m2 for Li and Sn, respectively) was still larger than/borderline to the power handling limit of the LMD concepts considered. Moreover, the impurity concentration in the pedestal - a proxy for the core plasma dilution/contamination - was computed to be above/close to tolerability limits suggested by previous COREDIV calculations. These results indicate that the operational window of an LMD for the EU DEMO, without any additional impurity seeding, might be too narrow, if it exists, and that Sn looks more promising than Li. A second set of calculations was then performed simulating Ar seeding in the SOL, to further reduce the target heat load, and consequently the metal erosion rate. It was found that the mitigation of the plasma heat load due to Ar radiation in the SOL effectively replaces the radiation associated to vapor shielding in front of the target, thus allowing to operate the LMD in a regime of low target erosion. The resulting operational window was found to be significantly wider, both in terms of tolerable peak target heat flux and of acceptable core plasma contamination.