Predicting the properties of multicomponent molten salts using density functional theory (DFT) remains challenging because the spatial and temporal scales required to evaluate transport properties and phase behavior are computationally prohibitive. In this work, we develop a moment tensor potential trained using a a DFT dataset of NaCl, KCl, NaCl-KCl mixtures, and the NaK alloy, enabling large-scale molecular dynamics simulations across wide ranges of temperatures and compositions. We systematically evaluate the effect of D3 dispersion corrections and apply the resulting potential to predict liquid densities, diffusion coefficients, radial distribution functions, heat capacities, thermal conductivities, and the NaCl-KCl phase diagram. The model successfully reproduces many temperature- and composition-dependent trends. However, systematic deviations in several absolute properties persist, highlighting the importance of experimental validation and calibration. These findings support a hybrid modeling framework in which first-principles-informed machine-learning potentials provide transferable predictive capability and mechanistic insight, while experimental data incorporated during model development or subsequent engineering assessments is necessary to improve quantitative accuracy.
The viscosity and density of organic mixtures are essential properties for designing lubricants, solvents, and heat transfer fluids. In engineering practice, formulating a functional fluid requires understanding how these properties change with composition and temperature. However, exhaustive experimental characterization across the full parameter space is impractical due to the vast number of possible species and combinations. Here we introduce a mixture-aware 3D molecular representation learning strategy, built upon a pre-trained molecular encoder, that jointly encodes component structures, mole fractions, and temperature to achieve accurate predictions for organic mixtures. Fine-tuning on publicly available datasets covering a wide range of binary organic mixtures yields test-set R2 values of 0.973 for dynamic viscosity and 0.996 for density, significantly outperforming traditional machine learning baselines. Beyond this overall accuracy, the model captures non-monotonic viscosity changes upon mixing, surpassing simple linear or logarithmic mixing rules. The architecture is extendable to ternary and multicomponent mixtures, as verified via preliminary experiments. Using this model, we quantitatively analyze how molecular structure-branching, cycloalkane, and aromatic rings-affects viscosity-temperature behavior, which benefits the design of lubricants with superior viscosity-temperature performance. Altogether, this work provides a practical, data-driven tool for mixture property prediction, accelerating the rational formulation of functional fluids in chemical engineering.
Entropy-stabilized oxides (ESOs) open access to vast multicomponent compositional spaces, but identifying promising candidates remains challenging because of the large number of possible mixtures and the need to assess their stability against competing phases. In this work, we develop a high-throughput computational framework to screen equimolar quinary ESOs in the NaCl structure type by combining density functional theory (DFT), special quasirandom structures (SQS), convex-hull thermodynamics, and supervised machine learning. A consistent reference database of binary and ternary ordered oxides, including disordered phases such as all binary cation combinations in the NaCl-type oxide, is first constructed using GGA and meta-GGA calculations. Quinary disordered phases are then described by SQS supercells and used to train machine-learning models that predict the distance to the convex hull and the corresponding stabilization temperature over the full set of 4368 possible equimolar quinary compositions generated from 16 cation species. Among the tested models, an optimized multilayer perceptron provides the best predictive performance, with a test error of about 4 kJ/mol, while requiring explicit DFT calculations for only about 10% of the quinary systems. Comparison with experimental synthesis tests and computed decomposition paths further shows that the approach captures the main stability trends and the dominant competing phases, although absolute stabilization temperatures remain affected by systematic thermodynamic approximations. These results establish an efficient route for the data-driven exploration of multicomponent oxides and provide practical guidance for the experimental search for new ESOs.
Sébastien Junier, C. Barreteau, David B'erardan et al.· Solid State Sciences· 0 citations
Molten fluoride salts are promising heat-transfer media for advanced molten salt reactors (MSRs), where reliable thermophysical property determination is critical for component design and safety. We present an integrated multiscale framework that couples machine-learning-driven atomistic simulations with finite-element (FE) modeling to predict the heat-transfer performance of FLiBe-based salts in a linear heat exchanger. At the atomistic scale, Moment Tensor Potentials (MTPs), actively trained on ab initio data, are developed for pure FLiBe (66-34 and 74-26 LiF-BeF2 mol%), FLiBe-LaF3, and FLiBe-UF4. These potentials are used in molecular dynamics simulations to obtain temperature- and composition-dependent transport properties (density, viscosity, thermal conductivity, and isobaric heat capacity), which are mapped as inputs to a three-dimensional FE model of the experimental thermal loop. The FE model with literature transport properties reproduces the experimental heat-transfer behavior of pure FLiBe to within 10% in the laminar regime and 18% in the transitional and turbulent regimes, validating the end-to-end pipeline for this composition. The same model with MTP-MD-derived transport properties systematically overestimates the heat-transfer coefficient by 25-28%, an offset consistent with the MTP-MD biases on thermal conductivity and viscosity. Applied to the ternary systems FLiBe-LaF3 and FLiBe-UF4 over 0-5 mol%, the MTP-MD-driven FE model predicts a mean reduction in heat-transfer efficiency of 8-11% relative to pure FLiBe, with UF4 exhibiting the strongest effect. The qualitative ordering of the three systems is the more robust result; the absolute value of the 8-11% figure is contingent on the MTP accuracy. The framework is complementary to high-temperature experiments and provides a physics-based pathway for the rapid screening of MSR coolant formulations.
M. Polovinkin, K. Abramova, O R Rahmanova et al.· 0 citations
Accurate prediction of natural gas thermophysical properties is essential for applications in production and transportation engineering, including reservoir simulation and flow modeling. Although machine learning (ML) techniques have been widely used, most studies focus on the estimation of these properties, with limited integration into practical applications. In this study, we propose a supervised model based on a Backpropagation Neural Network for simultaneous estimation of four interdependent properties: compressibility factor (Z), viscosity (μ), density (ρ) and gas formation volume factor (Bg). The multi-output model was trained on 58,165 data points generated from thermodynamic correlations, using pressure, temperature, composition (mole fractions of N2, CO2 and H2S), and gas specific gravity as inputs. The results yielded RMSE values of 5.56 × 10−4, 3.24 × 10−5, 3.01 × 10−2, and 6.33 × 10−4 for Z, μ, ρ and Bg, respectively, with R2 coefficients close to unity. The model’s applicability was evaluated by integrating the Z-factor into pressure drop calculations in pipelines using the Cullender and Smith method, resulting in a mean percentage error of 3.78%, close to the traditional method (3.83%). The results indicate that the model is an efficient and consistent alternative, highlighting the potential for integrating ML with classical hydraulic models.
Carolina Lima da Silva, L. C. Santos, G. Simonelli· Modelling· 0 citations
Accurate prediction of a material's melting temperature is critical for materials design and high-temperature applications. In this work, we investigate melting behavior across a deliberately selected set of elemental metals spanning systems where cohesive-energy trends suggest that PBE and PBEsol are expected to perform differently, as well as cases where their performance is ambiguous. Melting temperatures are computed using the two-phase coexistence (TPC) approach in conjunction with a machine-learned interatomic potential based on the moment tensor potential (MTP) framework, enabling large-scale simulations that minimize finite-size effects and ensure sufficient equilibration. The TPC-MTP results reveal a clear functional dependence in the predicted melting temperatures. PBE provides good agreement for several lighter elements, whereas PBEsol gives the best overall agreement across the full dataset. However, the element-resolved trends are not governed by cohesive energy alone, indicating that liquid-phase energetics, anharmonicity, and finite-temperature phase stability also contribute to the observed functional dependence. For intermediate and structurally complex systems, both functionals exhibit less systematic performance. Overall, this study provides a systematic assessment of functional-dependent melting temperature predictions, highlighting both the strengths and limitations of the combined TPC-MTP methodology and underscoring the need for carefully selected exchange-correlation treatments in high-accuracy melting-point simulations.
Pandu Wisesa, Christopher M. Andolina, W. Saidi· Journal of Chemical Physics· 0 citations
Sodium hydride (NaH) is the main by-product generated during the operation of the cold traps in a sodium-cooled fast reactor. Its thermal decomposition releases hydrogen gas, posing a safety hazard. However, understanding the atomic-scale decomposition mechanism remains challenging because conventional simulation methods are limited in the accessible length and time scales. This study developed a deep neural network potential (DP) for the NaH system using the DP-GEN active learning framework. Benchmark tests show that the DP model accurately reproduces density functional theory (DFT) reference energies, forces, equations of state, elastic properties, and phonon spectra. In particular, the DP-predicted bulk modulus and lattice constant are in good agreement with DFT results and close to experimental values, significantly outperforming the empirical ReaxFF potential. Subsequently, we conducted large-scale Deep Potential Molecular Dynamics (DPMD) simulations to investigate the thermal decomposition behavior of NaH clusters and a slab model. The simulation results reveal model-dependent thermal responses of NaH. In the original Na48H48 cluster simulation heated from 100 to 1200 K, a structural transition and disordering were observed, but no H2 formation occurred within the simulation time. In contrast, the slab model heated from 300 to 1500 K exhibited surface disordering, Na-H bond cleavage, H-H bond formation, cluster detachment, and H2 formation and release at elevated temperatures. A supplementary higher-temperature Na48H48 cluster simulation further showed cluster dissociation in the 1000–1500 K range. These results provide atomistic insight into the model- and temperature-dependent decomposition behavior of NaH and suggest that the DP model is a useful tool for studying hydrogen-related processes in alkali metal hydrides.
Ce Feng, Yuting Zhang, X. Zhang et al.· Materials· 0 citations