Intermediate-Range Order of Molten LiF-BeF2 and LiF-ThF4, Analyzed Via Molecular Dynamics Simulation and Network Topology Algorithms
Leslie Dewan, Massachusetts Institute of Technology
Molten salts have many applications in the nuclear power industry. Specifically, they can be used as both fuel and coolant in molten salt reactors, an innovative Generation IV nuclear reactor concept with numerous benefits in terms of safety and actinide management.
The temperature and ionic composition of the salts have a large influence on their structure and transport properties, which in turn have significant effects on a nuclear reactor’s neutronic and thermohydraulic behavior. Many structure and transport properties can be examined by experimental techniques, but the available data are currently sparse. Furthermore, it is difficult to measure some transport properties, such as diffusion coefficients, of these high-temperature, radioactive, and corrosive liquids in an experimental setting. Molecular dynamics (MD) simulations, however, are well-suited for evaluating these properties and can therefore be used successfully (generally in conjunction with experimental techniques) to fill in the gaps in existing experimental data.
This analysis uses MD potentials of ab initio accuracy that incorporate polarization effects to simulate the molten salts LiF-BeF2 and LiF-ThF4 at a range of temperatures and compositions. These potentials have been tested against experimental data, and are shown to be accurate.
We also use network topology algorithms to probe the intermediate-range order of these salts, which allow us to directly evaluate the connections between the salts’ structure and transport properties. These algorithms represent the system as an undirected, unweighted graph, in which each atom is a node and each bond is an edge. These topological algorithms enable a revealing assessment of the interrelationship between the molten salts’ structure and transport properties.
Abstract Author(s): Leslie Dewan, Mathieu Salanne, Linn Hobbs