Computational methods to simulate molten salt thermophysical properties.

Area of Science:

• Materials Science
• Chemical Engineering
• Computational Physics

Background:

• Molten salts are vital thermal conductors for molten salt reactors and solar thermal applications.
• Accurate thermophysical properties are essential for the safe and efficient use of molten salts.
• Experimental measurement of these properties across the full chemical space is challenging and often unfeasible.

Purpose of the Study:

• To review the evolution of computational methods for molten salt property prediction over the past 70 years.
• To identify the key advances and limitations of different simulation eras.
• To provide insights into future research directions, particularly machine learning.

Main Methods:

• Historical review and categorization of computational simulation methods into three distinct eras.
• Analysis of the strengths and weaknesses of methods from each era.
• Discussion of emerging trends, focusing on machine learning applications.

Main Results:

• The development of computational methods has progressed significantly over seven decades.
• Each era of simulation techniques presents unique advantages and inherent limitations.
• Machine learning is poised to become a dominant force in molten salt simulations.

Conclusions:

• Computational approaches are indispensable for overcoming experimental limitations in characterizing molten salts.
• Understanding the historical progression of methods aids in appreciating current challenges and future opportunities.
• The integration of machine learning promises to revolutionize the simulation of molten salt thermophysical properties.