Examination of the short-range structure of molten salts: ThF₄, UF₄, and related alkali actinide fluoride systems

Abstract

The short-range structures of LiF-ThF4, NaF-AnF4, KF-AnF4, and Cs-AnF4 (An = Th, U), were probed using in situ high temperature Extended X-ray Absorption Fine Structure (EXAFS) spectroscopy. Signally, the EXAFS...

A Holistic Approach for Elucidating Local Structure, Dynamics, and Speciation in Molten Salts with High Structural Disorder

“…Because of the inability to separate between multiple coordination states for Ni­(II) that coexist in molten ZnCl 2 , EXAFS fitting using Gaussian distributions fails to accurately describe the local coordination structure, drastically underestimating the coordination number (CN) of Ni­(II) to an unphysical value (CN < 2) . Therefore, adopting beyond-Gaussian modeling, which is possible via MD-EXAFS (a combination of molecular dynamics simulations and subsequent EXAFS calculations), is indispensable to analyze large and asymmetric structural disorder. − However, the success of this approach for complex mixtures with several competing coordination states is less certain. AIMD, which utilizes density functional theory (DFT) for treating interionic interactions, may predict which coordination structures are present in MS, − ,,, but sampling all relevant coordination structures rigorously with correct volume fractions (note that measured EXAFS corresponds to the volume average of the local structural motifs is not expected to be fully accurate due to uncertainty in the accuracy of DFT functionals in computing relative energies and the high computational cost of AIMD that limits simulations to short times (only tens to hundreds of picoseconds (ps)).…”

Modeling Metallic Halide Local Structures in Salt Melts Using a Genetic Algorithm

“…USPEX produced thousands of atomic configurations that represented well not only the lowest-energy solid structures but also structures that contained metal–halogen clusters in a large vacuum and single-metal atoms in highly concentrated alkali–halogen compositions. Our structural data set included many stoichiometric ratios similar to those used in MD calculations. ,, Therefore, it is not surprising that our results diverged from experimental measurements of room-temperature solids because our method uses probabilistic averaging of multiple structures in order to replicate the varying CNs in liquid environments, while pure solids have well-defined CNs. EXAFS fits of our most stable USPEX crystal structures showed that they resemble the observed local structures, allowing us to gain insights into the first and second nearest neighbors in the melt.…”

“…These experiments are established for systems with singular coordination environments or oxidation states. Still, they can be challenging for metals with multiple oxidation states and coordination environments. , Computational investigations can be used to decipher experimental investigations, providing a decomposition of the multiple coordination environments. ,,− …”

“…Molecular dynamics (MD) calculations coupled with XAS investigations have been frequently used to determine coordination states in salt melts. − , However, both classical MD potentials and density functional theory (DFT)-based methods have potential issues when modeling actinide salts. For example, classical potentials may not capture the complex interactions between actinide metals and typically do not describe the proper ordering that occurs in ionic liquids.…”

Experimental and Computational Exploration of the NaF–ThF4 Fuel System: Structure and Thermochemistry

“…Hence, dimers and all species of higher nuclearity were included in the model as dimers in which the overall coordination sphere around the two Th atoms is made of 13 fluorine atoms (the most abundant according to MD simulations), with the corresponding cation Th 2[XIII] 8+ . Given that pure ThF 4 (l) is a network of mostly corner-sharing coordination polyhedra , , it was modeled as a solution of dimers. Such a choice is a simplification, yet one which already reflects a main structural characteristic of the network: fluoride bridging.…”