Improved defect analysis based on atomic connectivity in polycrystalline materials.

Area of Science:

• Materials Science
• Computational Materials Science
• Condensed Matter Physics

Background:

• Material performance relies on microstructure, but degrades over time.
• Understanding defect generation is critical for high-temperature/energy applications due to rapid degradation and severe consequences.
• Current methods for lattice defect identification in atomistic simulations of polycrystals are unreliable, especially for non-ideal structures.

Purpose of the Study:

• To develop novel, reliable techniques for identifying and classifying point defects in atomistic simulations of polycrystalline materials.
• To address the limitations of conventional methods that rely on initial atomic positions and fail for inherently defective structures.

Main Methods:

• Introduced two new defect analysis techniques based on local atomic connectivity.
• Applied these methods to classify and quantify point defects in atomistic simulations.
• Validated the methods by comparing their performance against conventional techniques in collision-cascade simulations.

Main Results:

• The new methods accurately capture defect-production trends in collision-cascade simulations, unlike existing approaches.
• Demonstrated robust and accurate defect classification for polycrystalline materials, which possess inherent defects.
• The techniques are scalable and provide reliable defect identification based on actual lattice points.

Conclusions:

• The developed local atomic connectivity-based methods offer a significant advancement in defect analysis for polycrystalline materials.
• These scalable and accurate techniques overcome the long-standing challenge of reliable lattice defect identification in atomistic simulations.
• The findings are crucial for ensuring the safe operation and predicting the long-term behavior of materials in demanding applications.