Accurate and Efficient Calculation of the Solution Enthalpy and Diffusivity of Solutes in Liquid Metals Using Machine...

Area of Science:

• Materials Science
• Computational Chemistry
• Chemical Engineering

Background:

• Liquid metals (LMs) are crucial for energy systems, robotics, and electronics.
• Controlling LM chemical reactivity, influenced by solute stability and mobility, is key to enhancing their capabilities.
• Understanding solute behavior in LMs is essential for developing new materials and applications.

Purpose of the Study:

• To develop an automated, efficient method for calculating solute solution enthalpy and diffusivity in liquid metals.
• To leverage machine learning for predicting the chemical reactivity of liquid metals.
• To reduce computational costs associated with simulating solute behavior in LMs.

Main Methods:

• An automated method utilizing a machine learning moment tensor potential was developed.
• The method was tested using liquid sodium (Na) as a case study.
• Calculations focused on solution enthalpy and diffusivity of solutes in LMs.

Main Results:

• The machine learning method achieved accuracy comparable to first-principles molecular dynamics.
• Computational costs were reduced by a factor of 10 to 100.
• Demonstrated efficiency in predicting solute properties in liquid sodium.

Conclusions:

• The proposed automated method significantly accelerates the calculation of solute properties in LMs.
• This advancement is expected to foster progress in liquid metal chemistry and materials development.
• The method provides a pathway for better understanding and controlling LM reactivity for diverse applications.