Machine Learning Potential-Driven Investigation of NEPE Matrix: Mechanical Properties and Failure Mechanism

Area of Science:

• Materials Science
• Computational Chemistry
• Aerospace Engineering

Background:

• Mechanical properties of NEPE propellants are crucial for solid rocket engine safety and performance.
• Understanding NEPE failure mechanisms at atomic to micron scales is challenging.

Purpose of the Study:

• To apply machine learning potentials (MLP) for accurate and efficient simulation of NEPE mechanical properties.
• To investigate the influence of molecular size, strain rate, and temperature on NEPE's mechanical behavior.
• To validate simulation results with experimental data using the time-temperature superposition (TTS) principle.

Main Methods:

• Development and application of a machine learning potential (MLP) for NEPE.
• Molecular dynamics (MD) simulations to study NEPE under varying conditions.
• Utilizing the time-temperature superposition (TTS) principle to bridge simulation and experimental scales.

Main Results:

• MLP achieved ab initio-level accuracy with enhanced computational efficiency.
• NEPE's mechanical performance showed high sensitivity to temperature, with tensile strength decreasing significantly between 240-330 K.
• Predicted tensile strength (8-22 MPa) closely matched experimental data, validating the MLP-MD-TTS approach.

Conclusions:

• The study establishes a robust framework for simulating NEPE properties using MLP and multiscale modeling.
• Findings provide critical insights for optimizing NEPE safety and performance in solid rocket applications.
• This integrated approach advances high-performance propellant design.