Molten salts have attracted interest as potential heat carriers and/or fuel solvents in the development of new Gen IV nuclear reactor designs, high-temperature batteries, and thermal energy storage. In nuclear engineering, salts containing lithium fluoride-based compounds are of particular interest due to their ability to lower the melting points of mixtures and their compatibility with alloys. A machine learning potential (MLP) combined with a molecular dynamics study is performed on two popular molten salts, namely, LiF (50% Li) and FLiBe (66% LiF and 33% BeF 2 ), to predict the thermodynamic and transport properties, such as density, diffusion coefficients, thermal conductivity, electrical conductivity, and shear viscosity. Due to the large possibilities of atomic environments, we employ training using Deep Potential Smooth Edition (DPSE) neural networks to learn from large datasets of 141,278 structures with 70 atoms for LiF and 238,610 structures with 91 atoms for FLiBe molten salts. These networks are then deployed in fast molecular dynamics to predict the thermodynamic and transport properties that are only accessible at longer time scales and are otherwise difficult to calculate with classical potentials, ab initio molecular dynamics, or experiments. The prospect of this work is to provide guidance for future works to develop general MLPs for high-throughput thermophysical database generation for a wide spectrum of molten salts.

No abstract

Existing machine learning potentials for predicting phonon properties of crystals are typically limited on a material-to-material basis, primarily due to the exponential scaling of model complexity with the number of atomic species. We address this bottleneck with the developed Elemental Spatial Density Neural Network Force Field, namely Elemental-SDNNFF. The effectiveness and precision of our Elemental-SDNNFF approach are demonstrated on 11,866 full, half, and quaternary Heusler structures spanning 55 elements in the periodic table by prediction of complete phonon properties. Self-improvement schemes including active learning and data augmentation techniques provide an abundant 9.4 million atomic data for training. Deep insight into predicted ultralow lattice thermal conductivity (<1 Wm−1 K−1) of 774 Heusler structures is gained by p–d orbital hybridization analysis. Additionally, a class of two-band charge-2 Weyl points, referred to as “double Weyl points”, are found in 68% and 87% of 1662 half and 1550 quaternary Heuslers, respectively.

Thermal energy storage offers numerous benefits by reducing energy consumption and promoting the use of renewable energy sources. Thermal energy storage materials have been investigated for many decades with the aim of improving the overall efficiency of energy systems. However, finding solid materials that meet the requirement of high heat capacity has been a grand challenge for material scientists. Herewith, by training various machine learning models on 3377 high-quality data from full density functional theory (DFT) calculations, we efficiently search for potential materials with high heat capacity. We build four traditional machine learning models and two graph neural network models. Cross-comparison of the prediction performance and model accuracy was conducted among different models. The deeperGATGNN model exhibits high prediction accuracy and is used for predicting the heat capacity of 32,026 structures screened from the open quantum material database. We gain deep insight into the correlation between heat capacity and structure descriptors such as space group, prototype, lattice volume, atomic weight, etc. Twenty-two structures were predicted to possess high heat capacity, and the results were further validated with DFT calculations. We also identified one special structure, namely, MnIn2Se4, with space group no. 227 (Fd3̅m), that exhibits extremely high heat capacity, even higher than that of the Dulong–Petit limit at room temperature. This study paves the way for accelerating the discovery of novel thermal energy storage materials by combining machine learning with minimal DFT inquiry.

The discovery of new carbon allotropes with different building blocks and crystal symmetries has long been of great interest to broad materials science fields. Herein, we report several hundred new carbon allotropes predicted by the state-of-the-art RG2 code and first-principles calculations. The types of new carbon allotropes that were identified in this work span pure sp2, hybrid sp2/sp3, and pure sp3 C–C bonding. All structures were globally optimized at the first-principles level. The thermodynamic stability of some selected carbon allotropes was further validated by computing their phonon dispersions. The predicted carbon allotropes possess a broad range of Vickers’ hardness. This wide range of Vickers’ hardness is explained in detail in terms of both atomic descriptors such as density, volume per atom, packing fraction, and local potential energy throughout the unit cell, and global descriptors such as elastic modulus, shear modulus, and bulk modulus, universal anisotropy, Pugh’s ratio, and Poisson’s ratio. For the first time, we found strong correlation between Vickers’ hardness and average local potentials in the unit cell. This work provides deep insight into the identification of novel carbon materials with high Vickers’ hardness for modern applications in which ultrahigh hardness is desired. Moreover, the local potential averaged over the entire unit cell of an atomic structure, an easy-to-evaluate atomic descriptor, could serve as a new atomic descriptor for efficient screening of the mechanical properties of unexplored structures in future high-throughput computing and artificial-intelligence-accelerated materials discovery methods.