Machine learned interatomic potential for dispersion strengthened plasma facing components.

Area of Science:

• Materials Science
• Computational Materials Science
• Nuclear Engineering

Background:

• Tungsten (W) is ideal for fusion reactor divertors due to its high melting point and thermal conductivity.
• W's high brittle-to-ductile transition temperature and susceptibility to recrystallization at fusion temperatures (≥1000 K) necessitate improvements.
• Zirconium carbide (ZrC) dispersion strengthening can enhance W ductility and limit grain growth, but high-temperature effects are not fully understood.

Purpose of the Study:

• Develop a machine-learned interatomic potential for W-ZrC systems.
• Investigate the microstructural evolution and thermomechanical properties of W-ZrC at fusion reactor relevant temperatures.
• Understand the role of ZrC dispersoids in enhancing W ductility and high-temperature stability.

Main Methods:

• Trained a machine-learned Spectral Neighbor Analysis Potential (SNAP) on ab initio data for diverse W-ZrC structures, environments, and temperatures.
• Validated the potential's accuracy and stability using objective functions for material properties and high-temperature behavior.
• Performed atomistic simulations, including tensile tests on W/ZrC bicrystals at various temperatures.

Main Results:

• The W-ZrC SNAP potential accurately predicts lattice parameters, surface energies, bulk moduli, and thermal expansion.
• W(110)-ZrC(111) C-terminated bicrystals show high ultimate tensile strength (UTS) at room temperature, but strength decreases with increasing temperature.
• At 2500 K, carbon diffusion weakens the W-Zr interface in C-terminated samples, while Zr-terminated interfaces exhibit higher UTS.

Conclusions:

• The developed W-ZrC potential enables large-scale atomistic simulations at fusion reactor temperatures.
• Zr-terminated W(110)-ZrC(111) interfaces demonstrate superior high-temperature strength compared to C-terminated ones.
• ZrC dispersion strengthening offers a viable strategy to improve W's thermomechanical performance in fusion environments.