(335g) On the Onset of ‘Fuzz’ Formation in Plasma-Facing Materials: A Hierarchical Multiscale Modeling Approach

Fusion energy is the most promising carbon-free and sustainable energy source. However, harnessing fusion energy for the electrical grid is extremely challenging. One major problem that makes the realization of fusion energy particularly challenging is the severe surface degradation of plasma-facing-component (PFC) tungsten in linear plasma devices and fusion tokamaks. Tungsten is the PFC material for ITER (International Thermonuclear Experimental Reactor) and exhibits formation of nanometer-sized tendrils, or porous “fuzz”, following low-energy helium exposure. Helium (He) produced in the deuterium-tritium fusion reaction is a natural impurity in the plasma and it is extracted from the plasma through the divertor in ITER. Experimental studies have revealed that high particle fluxes of this low-energy helium produce fuzz-like nanofibers at the tungsten divertor within the temperature range from 900 K to 2000 K (consistent with the temperatures expected at the divertor wall due to the anticipated heat loads), which has adverse effects on the reactor performance.

Fuzz formation in plasma-facing tungsten is a complex multi-physics phenomenon resulting from the combined effects of driven surface diffusion, subsurface bubble dynamics, bubble bursting and surface crater formation, dislocation loop punching, anisotropies in material properties, changes in material thermophysical properties in the damaged tungsten, and many more. Developing a predictive model capable of capturing the formation and evolution of the fuzz-like complex surface morphology mediated by the underlying dynamical processes that are characterized by disparate spatiotemporal scales has been a major theoretical and computational challenge. While large-scale molecular-dynamics (MD) simulations have been used to successfully identify many of the aforementioned phenomena [1], simulating experimentally the spatiotemporal scales (mm-hr) of relevance to fuzz formation is well beyond the scope of such atomistic simulations. Therefore, we followed a hierarchical multiscale modeling paradigm to develop a continuum-scale model for simulating the onset of fuzz formation in He plasma-irradiated tungsten [2].

In our modeling framework, large-scale MD simulation results are used to parameterize the constitutive equations required for the closure of the continuum-scale model and targeted MD simulations are used to determine the thermophysical properties [3] of He-implanted tungsten. One highlight of this framework is the calculation of the bubble-matrix (He-W) interfacial free energy, a key parameter in the equation of state (EOS) for the He density in the bubble [4]. We have employed a novel iterative scheme, employing a protocol of MD simulations and the Young-Laplace equation, to solve the nonlinear EOS and determine the value of the bubble-matrix interfacial free energy. Furthermore, we have used a cluster-dynamics model to simulate the subsurface helium bubble evolution and bubble bursting and develop a coarse-grained model for He concentration evolution in plasma-facing tungsten, which can be coupled with the surface evolution model. Our cluster-dynamics model, implemented numerically in our code Xolotl, has been parameterized based on large-scale MD simulation results and benchmarked against experimental measurements [5]. Based on this atomistically-informed continuum-scale model, we have conducted self-consistent numerical simulations of the He-implanted tungsten surface morphological evolution and validated the model by comparing the simulation predictions with experimental measurements.

[1] K. D. Hammond, S. Blondel, L. Hu, D. Maroudas, and B. D. Wirth, Acta Mater. 144, 561-578 (2018).

[2] D. Dasgupta, R. D. Kolasinski, R. W. Friddle, L. Du, D. Maroudas, and B. D. Wirth, Nucl. Fusion 59, 086057 (2019).

[3] A. Weerasinghe, B. D. Wirth, and D. Maroudas, ACS Appl. Mater. Interfaces 12, 22287–22297 (2020).

[4] K. D. Hammond, D. Maroudas, and B. D. Wirth, Sci. Rep. 10, 2192 (2020).

[5] S. Blondel, D. E. Bernholdt, K. D. Hammond, and B. D. Wirth, Nucl. Fusion 58, 126034 (2018).