Predicting impurity profiles in the Infinity Two optimized stellarator fusion pilot plant

Significant accumulation of impurities in the core of magnetically confined fusion plasmas must be avoided to limit fuel dilution or excessive radiation that can degrade fusion performance. Following initial transport predictions in the Infinity Two fusion pilot plant [1], we perform additional analysis to predict impurity profile peaking. The analysis is performed for He ash generated as the product of D-T fusion, W atoms potentially sourced in the plasma by sputtering of metallic walls, and Ne atoms seeded in the divertor to increase radiation fraction for exhaust component compatibility. The particle flux is calculated for both neoclassical transport using the SFINCS code [2] and nonlinear turbulent transport using the GX code [3]. Parameter scans in the impurities’ density and temperature gradients are run in the trace limit to calculate the diffusive and convective coefficients in the presence of the main plasma. Turbulent transport is found to dominate neoclassical contributions everywhere in the plasma, ensuring the avoidance of potentially catastrophic neoclassical pinch of high-Z impurities. In the deep-core with a flat main ion density, weak peaking of all impurity species is predicted due to ion temperature gradient (ITG) turbulence. In the outer region where there is large main ion density gradient from pellet fueling, the impurities exhibit a stronger species-dependent convection due to trapped electron mode (TEM) turbulence. Here, the He exhibits an inward pinch, while Ne and W exhibit increasingly larger outward convection. Using predicted transport coefficients at multiple radii, the integrated Ne and W (source-free) density profiles are predicted to be flat and hollow, respectively. Assuming the D-T fusion reactivity rate as a thermal He source profile, the resulting He density profile is peaked. However, all impurity profiles are predicted to have smaller peaking, and therefore smaller core concentrations and Zeff, than assumed in initial performance projections [1] confirming desired performance will be maintained or improved. Additional calculations are performed to quantify the impact of using species concentrations in the non-trace limit.

[1] W. Guttenfelder, et al. 2025 J. Plasma Phys.

[2] M. Landreman, et al. 2014 Phys. Plasmas.

[3] N. R. Mandell, et al. 2018. J. Plasma Phys.