Session: 04-01-01: Advanced Materials for Energy

Paper Number: 150983

150983 - Structure and Dynamics of Uranium-Chlorine Complexes in Uranium Chloride - (Kcl-Licl) Molten Salts – an Atomistic Perspective 

Molten salt reactors (MSRs) are promising technology for nuclear reactor designs owing to their inherent safety features, high thermal efficiency, and potential for sustainable energy generation. These reactors frequently employ molten chloride salts to dissolve nuclear fuels due to their high boiling points, wide liquid temperature ranges and low vapor pressure. The structure and thermophysical properties of these molten salts, which depend on their composition and the oxidation state of the uranium, is not well understood. The goal of this study is to evaluate the structure of UCl_n (n = 3,4) – (KCl - LiCl) eutectic at various compositions and oxidation states. The study delves into the molecular-level interactions, elucidating the coordination environments, solvation shell rigidity, and dynamics in uranium-chlorine complexes.

Using classical and ab initio molecular dynamics, we simulated a range of salt compositions and analyzed the diads and triads in U – U forming chemistry. We initially validated our AIMD calculations for 16.25 wt% UCl₃ in eutectic (KCl-LiCl) mixture. This validation attested our pseudopotentials and basis sets used for the AIMD calculations with CP2K software. The AIMD simulations employed the PBE functionals and the energy was evaluated using a 1 × 1 × 1 k-point mesh. The wave functions of valence electrons, Li(2s¹), Cl(3s² 3p⁵), K_sv(3s² 3p⁶ 4s¹), and U_s(5f³ 6s² 6p⁶ 6d¹ 7s²), were expanded in the plane wave basis set with a cutoff energy of 420 eV, and the core electrons were approximated by projector augmented wave (PAW) pseudopotentials. Spin polarization was always applied for AIMD simulations to properly depict unpaired electrons of U³⁺. The structure and density obtained from these simulations matched extremely well with prior published data. In particular, the results indicate a 6-fold distorted octahedral U-Cl coordination structure.

While AIMD simulations were quite accurate in modeling the short-range structure, they are limited to systems comprising ~100 atoms and can access ~20 ps simulation time. On the other hand, classical molecular dynamics based on empirical potentials are able to scale to much larger systems and access much longer simulation times to model diffusive behavior at a much larger length scale. Therefore, in this study we employed classical MD with Born-Meyer-Huggins potential, which had been previously used to calculate thermodynamic properties of low concentration UCl₃ molten salts in eutectic (KCl - LiCl). These simulations enabled us to study much larger systems with higher concentration of UCl_n. For instance, we simulated 50 mol% UCl_n(n =3,4) in KCl – LiCl eutectic at 1100 K. The results suggest 6 and 7-fold coordinated structures in the solution environment for U³⁺-Cl and U⁴⁺-Cl respectively, which influences the ion transport dynamics of Uⁿ⁺ (n = 3,4) in the (KCl - LiCl) eutectic. The results from the calculations, such as solvation enthalpies, were compared with experimental calorimetric measurements and provide crucial inputs to perform a detailed thermodynamic analysis of the salt-complexes. Overall, the study addressed specific challenges in modeling and predicting uranium-chlorine coordination behavior and highlighted the need for multiscale approach to address complex salt-fuel interactions that play a crucial role in determining thermophysical properties of molten salts and therefore influence MSR design.

Presenting Author: Soumik Banerjee Washington State University

Presenting Author Biography: Dr. Banerjee is an Associate Professor in the School of Mechanical and Materials Engineering (MME) at Washington State University and currently serves as the Graduate Program Chair at the School of Mechanical and Materials Engineering. He also serves as an Associate Editor of the Journal of Electrochemical Energy Conversion and Storage as well as on the editorial board for Computational Thermal Sciences. Dr. Banerjee’s research expertise lies in modeling of processing, structure and functional properties of materials and interfaces relevant to energy conversion and storage. Dr. Banerjee routinely employs ab initio quantum mechanical calculations, atomistic and molecular modeling techniques, as well as stochastic models such as kinetic Monte Carlo to simulate a range of materials including electronic materials, ceramics, hybrid perovskites, sulfide glasses, and ionic liquids. Dr. Banerjee is an elected Fellow of the American Society of Mechanical Engineers (ASME) and has received several prestigious awards including the 3M Non-tenured Faculty Award in 2013 and the Pratt Fellowship at Virginia Tech. He has published over 50 peer-reviewed articles and presented more than 50 times at national and international meetings. Dr. Banerjee’s work has been widely cited by others and his scholarly work has an h-index of 20 and i-10 index of 28. Dr. Banerjee has been active in his scientific community and has organized symposia, topics and sessions at several conferences.

Authors:

Azmain Islam Washington State University
Xiaofeng Guo Washington State University
Soumik Banerjee Washington State University