Combined DFT and ionic model computational study of the effect of alkali metal cations on the structure, bonding and properties of molten cryolites, M₃AlF₆ (M = li, Na, K, Rb, and Cs)

ABSTRACT

Ionic melts, in the form of ionic liquids and molten salts, are important in many technological aspects such as metal production and energy applications. Cryolite melts constitute an important family especially for the industrial production of Al. The present computational study aims to analyse the effect of different cations on the structure and properties of Al (III) fluorocomplexes in cryolite melts. Cation effects are discussed as the ion pair interactions due to alkali metal cations of cryolite compounds M₃AlF₆ with M= Li, Na, K, Rb, or Cs for various microclusters using ionic interaction model (IM) and DFT calculations. DFT results for different ion complexes are compared with the energy calculations using a pseudoclassical model for isolated clusters. QTAIM and ELF analyses were performed to understand the nature of interactions in detail. A link between open-shell interactions and decreased ionic conductivity is proposed within the series of K, Rb, and Cs. Environmental effects, such as temperature and presence of other ions, are investigated in detail by DFT analysis.

KEYWORDS:

• Molten salts
• ionic melts
• cryolite melts
• cation effect
• DFT
• ionic model
• QTAIM

Acknowledgments

Authors gratefully acknowledge also the Computational Materials Science Laboratory of Piri Reis University for computational resources.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supporting information

Relative energies in vacuo, relative enthalpies and entropies both in vacuo and in continuum, Dipole moments, ${\mathrm{\nabla }}^2\mathit{\rho }\left(\vec{r}\right)$ maps, and selected QTAIM parameters for four, five, and sixfold coordinations of M₃AlF₆ with M=Li, Na, K, Rb, Cs; HOMO plots for fourfold structures are available as Supporting Information.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/00319104.2024.2349777

Additional information

Funding

Computing resources used in this work were provided by the National Center for High Performance Computing of Turkey (UHeM) under grant number [1017342023]. The numerical calculations reported in this paper were partially performed at TUBITAK ULAKBIM, High Performance and Grid Computing Center (TRUBA resources).