Mechanism-Aware Digital Twin for High-Temperature Creep Prediction in Mo-Re Alloys

Area of Science:

• Materials Science
• Mechanical Engineering
• Computational Materials Science

Background:

• Predicting long-term deformation in structural materials at extreme temperatures is challenging.
• High-temperature creep significantly limits the performance and lifespan of refractory alloys.
• Understanding creep mechanisms is crucial for designing advanced materials.

Purpose of the Study:

• To develop a physics-informed digital twin for predicting and understanding the creep behavior of Mo-14Re alloys.
• To identify novel creep mechanisms in refractory alloys under extreme conditions.
• To demonstrate a transferable AI-enabled approach for materials design.

Main Methods:

• Integration of a viscoplastic self-consistent (VPSC) model with real-time creep experiments.
• Utilizing a calibration neural network within a digital twin framework.
• Employing transmission electron microscopy (TEM) and molecular dynamics simulations for validation.

Main Results:

• The digital twin accurately reproduced experimental creep curves (1000-1200 °C, 60-150 MPa) with <5% deviation.
• A previously unrecognized solute-drag mechanism involving rhenium segregation at grain boundaries was identified.
• Nanoscale La2O3 precipitates were shown to suppress rhenium segregation and improve creep resistance.

Conclusions:

• The study advances the mechanistic understanding of refractory alloy creep behavior.
• The developed AI-enabled digital twin framework is transferable for designing materials under extreme environments.
• The findings offer insights into optimizing refractory alloy performance for high-temperature applications.