Machine learning has transformed materials simulation by delivering force fields with ab initio accuracy, yet bridging the gap between high-dimensional regression and physical interpretability remains a grand challenge. Conventional analytical potentials offer transparency but often fail to capture the complexity of far-from-ground-state regimes. Here, we introduce a hybrid symbolic–neural framework that unifies the interpretability of the embedded atom method (EAM) with the adaptability of data-driven learning. Using Equation Neural Networks (EqNNs) trained on density functional theory (DFT) data, we obtain interpretable models for aluminum through three distinct training protocols: random initialization trained via Monte Carlo Tree Search (MCTS) and gradient descent, and two transfer-learning strategies initialized from a copper potential—one employing MCTS followed by gradient descent and the other using gradient descent only. We find that although all three resulting symbolic models achieve sub-10 meV/atom accuracy, they occupy distinct local minima in the functional landscape, exhibiting complementary trade-offs across phonon dispersion, surface energetics, and elastic response. By integrating these diverse functional forms through a weighted symbolic ensemble, we derive a composite potential that surpasses the fidelity of its constituent models. The resulting ensemble effectively mitigates individual biases, delivering superior consistency with DFT benchmarks across equation-of-state curvature, phonon spectra, and melting dynamics. This approach demonstrates that combining transfer learning with ensemble symbolic regression yields compact, transparent potentials capable of robust prediction across equilibrium and nonequilibrium states.

Sodium-ion batteries employing hard carbon electrodes are considered a drop-in technology for lithium-ion batteries. Electrode drying is a critical manufacturing step, as binder migration during pore emptying impacts the mechanical integrity and electrical performance of the electrode. Existing modeling approaches predominantly rely on the film shrinkage phase in a one dimensional way or neglect the capillary transport, resulting in a lack of physically consistent microstructure resolved predictions of binder migration. In this work, a spatially resolved pore scale continuum model is extended to explicitly describe capillary-driven binder transport during pore emptying. The model is applied to hard carbon microstructures with varying mean particle diameters. The simulations reveal that smaller particle sizes lead to a more homogeneous binder distribution, whereas higher evaporation rates and increased surface tension promote stronger binder gradients. Variations in solvent viscosity show only a minor influence on binder migration, as long as no hydrophilic or hydrophobic behavior is present. Finally, the simulations demonstrate that an explicit description of capillary transport and microstructural effects is essential for accurately predicting binder migration and provides a basis for the targeted optimization of electrode drying processes.

In GaAs/InAs core/shell nanowires, comprising a tubular conducting shell, interference phenomena observed under an axial field and originating from closed-loop states encircling the insulating core provide an ideal platform for superconducting quantum devices that utilize effects such as Aharonov-Bohm–type or Altshuler-Aronov-Spivak–type conductance oscillations. Both effects are different in nature with respect to phase rigidity because of interference of non-time-reversed or time-reversed paths, respectively. Since their occurrence is largely governed by averaging effects, which depend on sample dimensions and the transport regime, we present a systematic study of flux-periodic oscillations of phase-pure zinc-blende GaAs/InAs core/shell nanowires as a function of gate voltage for samples with different contact separation lengths. Our analysis shows that with increasing contact separation length, averaging effects result in gradual reduction of $h/e$-periodic Aharonov-Bohm–type oscillations, while relative to that the $h/2e$-periodic Altshuler-Aronov-Spivak oscillations and its $h/4e$-periodic higher harmonics are enhanced. The additional phase rigidity seen in the $h/3e$-periodic oscillations is attributed to phase rigidity propagating from the neighboring lower harmonics. Our tight-binding transport simulations on nanowires of different lengths which contain only a few scattering centers confirm the experimental observations regarding the different harmonics and their phase rigidity. Together, our experimental and simulation findings indicate quasiballistic transport with persistent Aharonov-Bohm–type and phase-rigid Altshuler-Aronov-Spivak–type oscillations despite few scattering centers.

Quantum zero-point effects have been long proposed to explain a well-known discrepancy between the low-temperature flow stresses of body-centered cubic metals and corresponding atomistic models of plastic flow. Previous investigations on quantum effects relied on empirical interatomic potentials, which poorly reproduce dislocation energy landscapes compared to density functional theory (DFT) calculations. Here, we revisit this problem using DFT and machine-learning interatomic potentials (MLIPs). We show that while quantum effects do contribute to dislocation glide at low temperature, their magnitude is much lower than previously reported, and insufficient to reconcile atomistic predictions with experiments. Our results thus reopen a long-standing question and challenge for predictive atomistic modeling, on a fundamental property of crystals.

Pretrained machine-learned interatomic potentials (MLIPs) lack tailored accuracy for specific material systems like molybdenum-oxygen (Mo-O) interactions. Herein, we develop a fine-tuned MLIP by optimizing the DPA-2 model with around one thousand data points from density functional theory (DFT) calculations to study O adsorption on Mo surfaces. The fine-tuned model achieves DFT-level accuracy in predicting key properties of Mo-O systems, including lattice constants, elastic moduli, and formation energies of bulk Mo and Mo oxides $({\mathrm{MoO}}_2$, ${\mathrm{MoO}}_3)$, as well as surface energies and O adsorption energies on Mo low-index surfaces across various coverages and sites. In contrast, the zero-shot DPA-2 model without fine-tuning exhibits substantial errors in complex nonbulk configurations, with errors in predicted energies, atomic forces, and virial tensors that are up to two orders of magnitude larger than those of the fine-tuned model when benchmarked against DFT results. Fine-tuning substantially reduces deviations from DFT, demonstrating that universal MLIPs can be effectively enhanced with small specialized datasets at low computational cost. This work provides a reliable tool for Mo-O surface simulations and offers insights for fine-tuning MLIPs in targeted material research.

The nonlinear Hall effect (NLHE) holds significant promise for next-generation electronic device applications, and it is essential to explore and engineer a strong and highly tunable NLHE. In this work, we report large second-order and third-order NLHEs in dual-gated ${\mathrm{BiSbTeSe}}_2$ devices. Both the second-order and third-order NLHE are highly gate tunable. The second-order NLHE exhibits a higher generation efficiency than those reported in previous topological insulator systems, and it can be stably observed up to room temperature. Meanwhile, the generation efficiency of the third-order NLHE is comparable to previously reported values. The scaling analysis indicates that the second-order NLHE is consistent with contributions from skew scattering and a term compatible with either the Berry curvature dipole or side jump, whereas the third-order NLHE is consistent with contributions from the Berry-connection polarizability tensor and higher-order skew scattering. Our work establishes a promising experimental platform for the development of NLHE-based devices.

${\mathrm{YbMn}}_6{\mathrm{Ge}}_6$ is an antiferromagnet in which ${\mathrm{Yb}}^{3+}$ moments couple to the kagome Mn sublattice, stabilizing a conical incommensurate magnetic state at low temperature. However, previously reported magnetization and resistivity behaviors remain partially inconsistent. Here, we demonstrate pronounced sample-dependent thermal and magnetic hysteresis in ${\mathrm{YbMn}}_6{\mathrm{Ge}}_6$ single crystals grown by the Sn-flux method and show that these variations are consistent with Yb-site deficiency and possible spatial inhomogeneity. Both structural and chemical composition analyses indicate Yb vacancies in all crystals. The transition onset temperature and its width vary between batches with different inferred Yb occupancies: in resistivity measurements, crystals with higher Yb occupancy exhibit transitions at 68 K (cooling) and 80 K (heating), whereas samples with lower inferred Yb occupancy display a broadened transition window of 110–128 K. While these transitions are also reflected in magnetization measurements, measurements on crystals and smaller fragments suggest that the apparent onset is governed not simply by macroscopic sample size but also by a distribution of local transition temperatures. Weak and sample-dependent signatures, including the presence of thermal hysteresis, are also found in specific heat measurements, indicating that spatial averaging over inhomogeneous regions can smear thermodynamic signatures. At low temperature, we further observe strongly anisotropic field-induced metamagnetic hysteresis and its sensitivity to the sample-dependent defect landscape. Magnetoresistance measurements track the metamagnetic features, indicating a close correlation between magnetic order and charge transport. Overall, our results suggest that Yb-site deficiency and associated spatial inhomogeneity are important factors governing the sample-dependent hysteretic behavior and highlight the importance of the R-site chemistry in the isostructural $R{\mathrm{Mn}}_6{\mathrm{Ge}}_6$ family.

The interlamellar space in van der Waals materials has served as a test bed for studies of strongly confined fluid and ion transport. Despite theoretical and experimental progress, it remains unclear how electrostatic correlations arising from interactions between the confined, hydrated ions and the substrate layer can be reconciled with nonequilibrium ionic transport and the substrate's morphological response. Here, we investigate field-driven ionic conduction in sodium- and water-intercalated layered manganese oxides $\left({\mathrm{MnO}}_2\right)$, a self-intercalated metal oxide nanofluidic system, as a model intrinsically intercalated van der Waals solid, using nonequilibrium all-atom molecular dynamics simulations that explicitly capture ion-water correlations and layer morphology. We demonstrate that electric-field bias induces spontaneous nanoscale segregation of water within the interlayer space, producing coexisting hydrated and weakly hydrated ionic domains coupled to local lattice distortions. This feedback between hydration and morphology gives rise to heterogeneous transport pathways and leads to a nonmonotonic dependence of ionic conductivity on water content. Suppressed conduction arises either from layer collapse at low water levels or from ionic hop blockage by excess water at high water levels. Concurrently, ionic transport exhibits a maximum at intermediate levels of intercalated water, where less hydrated ions can move at the boundaries of strongly hydrated ion clusters. While such nonmonotonicity could explain the experimentally observed memristive response of ${\mathrm{MnO}}_2$ single crystals, these findings also provide a molecular-level mechanism linking intercalated water levels to ionic metal oxide nanofluidic systems, suggesting general design principles for robust, water-assisted ionic conductors.

$3d-5d$ transition-metal double perovskites exhibit diverse magnetic and electronic phases arising from the interplay of electron correlation ($U$) and spin-orbit coupling (SOC). Using first-principles calculations, we investigate the contrasting magnetic and charge states of ${\mathrm{Ba}}_{2}X{\mathrm{IrO}}_6(X=\mathrm{Ni},\mathrm{Co},\mathrm{Fe})$, incorporating the combined effects of $U$ and SOC. Structural and magnetic ground-state searches show that all compounds adopt a rocksalt structure, with ferromagnetic ordering in ${\mathrm{Ba}}_2{\mathrm{FeIrO}}_6$ and distinct antiferromagnetic phases (AFM-1 and AFM-2) in ${\mathrm{Ba}}_2{\mathrm{NiIrO}}_6$ and ${\mathrm{Ba}}_2{\mathrm{CoIrO}}_6$, respectively. Dynamical mean-field theory (DMFT) captures dynamic correlations and confirms nominal Ir charge states of $+6(X=\mathrm{Ni},\mathrm{Co})$ and $+5(X=\mathrm{Fe})$, leading to half-metallic (Fe), semiconducting (Ni), and metallic (Co) behavior. DMFT spectral functions corroborate the transport characteristics, while impurity charge analysis clarifies Ir charge variation. Exchange interactions within a classical Heisenberg model reveal ferromagnetic nearest-neighbor ($J_{X-\mathrm{Ir}}$) and antiferromagnetic next-nearest-neighbor ($J_{X-X}, J_{\mathrm{Ir}-\mathrm{Ir}}{J}_{\mathrm{Ir}-\mathrm{Ir}}$) couplings, with the latter stabilizing AFM order in the Ni and Co compounds. Benchmarking against available experimental data for ${\mathrm{Ba}}_2{\mathrm{NiIrO}}_6$ provides a comprehensive framework for exploring related double perovskites and guide future experimental studies.

Pump-probe optical methods such as transient grating spectroscopy (TGS) enable rapid, nondestructive thermoelastic property measurements. But, in phonon-dominated ceramics, they can underpredict bulk thermal diffusivity when long mean free path phonons do not equilibrate over experimental length scales. We combine in situ TGS with ${\mathrm{Si}}^{4+}$ ion irradiation of chemical vapor deposited 3C-SiC (300 and $550{}^{\circ }\mathrm{C}$, 0.5–1 dpa) and density functional theory informed Boltzmann transport equation solutions to understand the origins of this offset. We show how the discrepancy between laser flash analysis (LFA) and TGS-measured thermal diffusivity varies with grain-boundary density, temperature, and defect concentration. We introduce a dimensionless suppression factor that accounts for this discrepancy, and we demonstrate its utility by using it to show an agreement between the thermal defect resistance of neutron irradiated 3C-SiC (measured using LFA) and ion irradiated 3C-SiC (measured using TGS). This theory-informed experimental framework enables quantitative, in situ tracking of ion irradiation damage induced thermal transport degradation in ceramics.

Recently, transition metal dihalides have emerged as a promising class of two-dimensional materials with unique electronic and optoelectronic properties, although their defect-mediated surface chemistry remains largely unexplored. Herein, we employed density functional theory calculations to investigate the adsorption of gas molecules ($\mathrm{CO}, {\mathrm{CO}}_2, {\mathrm{H}}_2, {\mathrm{H}}_2\mathrm{O}, {\mathrm{N}}_2$, and ${\mathrm{O}}_2$) on pristine and defective $\text{2H-}{\mathrm{TiBr}}_2$ monolayers containing a $\mathrm{Br}$ monovacancy. The pristine substrate exhibits weak physisorption dominated by van der Waals interactions, preserving both the intrinsic semiconducting character and the electronic structure of the monolayer. In contrast, the introduction of a $\mathrm{Br}$ vacancy substantially enhances the local surface reactivity by exposing undercoordinated $\mathrm{Ti}$ atoms that act as active adsorption centers. While $\mathrm{CO}, {\mathrm{CO}}_2$, and ${\mathrm{H}}_2$ remain weakly adsorbed even in the presence of the vacancy, ${\mathrm{H}}_2\mathrm{O}$ exhibits a stronger molecular adsorption without bond dissociation, whereas ${\mathrm{N}}_2$ and ${\mathrm{O}}_2$ undergo strong chemisorption at the defect site. Combined Bader, PDOS, charge-density-difference, and structural analyses support a defect-mediated charge-transfer mechanism in which electrons are transferred from localized $\mathrm{Ti} d$ states toward low-lying molecular ${\pi }^{*}$ antibonding orbitals, weakening the intramolecular bonds and leading to significant $\mathrm{N}–\mathrm{N}$ bond elongation and dissociative adsorption of ${\mathrm{O}}_2$. The vacancy-induced functionalization also modifies the electronic structure by introducing localized midgap states, qualitatively reducing the PBE band gap, and stabilizing a magnetic moment of $1.00{\mu }_{\mathrm{B}}$. Our results show that defect engineering is a promising strategy to tune the reactivity and electronic properties of ${\mathrm{TiBr}}_2$ monolayers, providing an atomistic foundation for future studies on gas sensing and heterogeneous catalysis.

Epitaxial strain in perovskite oxide thin films is conventionally understood through lattice parameter modulation, characterized by the $c/a$ ratio of the pseudocubic cell. However, internal atomic arrangements such as octahedral tilting can relax independently, a phenomenon that has remained largely unexplored. Here, we demonstrate that electrochemical operation induces internal structural relaxation in epitaxial ${\mathrm{La}}_{0.7}{\mathrm{Sr}}_{0.3}{\mathrm{CoO}}_3$ films while in-plane lattice parameters remain pinned to the substrate. Synchrotron x-ray diffraction reveals that prolonged electrochemical treatment drives the octahedral tilt mode from the as-grown strained configuration toward the bulk structure without conventional lattice parameter changes. This relaxation is enabled by ionic oxygen migration during electrochemical reactions, which reduces energy barriers between structural configurations. Our findings reveal a relaxation pathway in epitaxial oxides that operates independently of conventional lattice parameter relaxation, and demonstrate the importance of monitoring internal atomic arrangements under operating conditions. This establishes electrochemically driven ionic migration as a versatile tool for controlling internal structure in epitaxial oxides, opening opportunities for designing strain-engineered functional materials.

Two-dimensional magnetic materials with tunable electronic and magnetic properties show great potential for spintronic applications. Using first-principles calculations, we systematically investigate monolayer ${\mathrm{CuPdBr}}_3$ and identify it as a zigzag antiferromagnetic (AF-Zigzag) semiconductor with an indirect band gap. While the AF-Zigzag ground state remains robust under strain, its magnetic configuration can be effectively tuned through carrier doping. Notably, both electron and hole doping induce a magnetic phase transition from the AF-Zigzag to a ferromagnetic (FM) state, with the stability of the FM state enhanced by increasing doping concentration. The corresponding critical temperature can be increased to about 100 K. Remarkably, antiferromagnetic (AFM) monolayer ${\mathrm{CuPdBr}}_3$, driven by carrier doping, behaves as a bipolar magnetic semiconductor (BMS). In the FM state induced by hole doping, the spin-up channel becomes metallic, giving rise to a ferromagnetic half-metal (FMHM), whereas electron doping results in a FMHM with a metallic spin-down channel. The carrier-doping-induced BMS behavior of AFM monolayer ${\mathrm{CuPdBr}}_3$ suggests its potential application in spin field-effect transistor devices. Furthermore, in carrier-doped monolayer ${\mathrm{CuPdBr}}_3$ with the FM state, spin-orbit coupling opens energy gaps around the spin-polarized non-Dirac band degeneracies at the $\mathrm{\Gamma }$ point, resulting in significant anomalous Hall conductivity. These findings highlight AFM monolayer ${\mathrm{CuPdBr}}_3$ as a promising candidate for the development of two-dimensional antiferromagnet-based spintronic devices and the exploration of novel magnetic phenomena.

Porous materials such as metal-organic frameworks (MOFs) are widely studied for applications in catalysis and gas adsorption, including carbon dioxide capture and hydrogen storage for energy and environmental challenges. Accurate simulation of gas adsorption in these materials remains challenging because most approaches rely on force fields (FFs), which lack first-principles accuracy. Here, we present a hybrid sampling approach, complemented by an active-learning-style refinement step, for generating machine-learning potentials (MLPs) tailored for grand canonical Monte Carlo (GCMC) simulations of gas adsorption. The training dataset combines snapshots from FF-based GCMC sampling, evaluated by single-point density functional theory (DFT) calculations, with trajectories from DFT-based molecular dynamics (MD). We demonstrate this approach for hydrogen adsorption in MOF-303, achieving accurate energy and force evaluation, as well as robust GCMC simulation including on-the-fly quantum corrections for nuclei. The MLP-based calculation also revealed four primary and four secondary adsorption sites formed near N, O, C, and –OH motifs, highlighting the material's structural heterogeneity and adsorption behavior. The resulting MLPs enable calculation of adsorption isotherms, isosteric heat values, and diffusion coefficients across temperatures. This hybrid-sampling framework bridges DFT-level accuracy and larger-scale simulation capability than $abinitio$ MD, and can in principle be applied to other porous materials and gas types to support the design of materials for gas storage and separation.

Hydriding is a primary degradation mode of plutonium metal, and although it tends to be mitigated by the presence of the native surface oxide, the degree of protection depends heavily on the oxide chemistry. Here, we use hybrid density-functional theory calculations to evaluate the thermodynamics and kinetics of hydrogen in ${\mathrm{PuO}}_2$ as a function of environmental oxygen content. Under O-rich conditions, hydrogen will incorporate primarily through interstitial protons ${\mathrm{H}}_i^{+}$, which peak in concentration at ${10}^{16}$ ${\mathrm{cm}}^{-3}$. ${\mathrm{H}}_i^{+}$ has very small kinetic barriers ($<0.1\mathrm{eV}$), although its diffusion will be limited by its small concentrations. For more O-poor conditions, hydrogen prefers to substitute for oxygen as ${\mathrm{H}}_{\mathrm{O}}$; this species can reach concentrations exceeding ${10}^{20}$ ${\mathrm{cm}}^{-3}$. ${\mathrm{H}}_{\mathrm{O}}$ diffuses into neighboring oxygen vacancies $V_{\mathrm{O}}$ with barriers between 0.39–0.71 eV, depending on the charge state of $V_{\mathrm{O}}$. These barriers will be overcome readily at room temperature. In addition, as conditions become increasingly O-poor, electron polarons ${\eta }^{-}$ will form in higher concentrations, reflecting reduction from ${\mathrm{Pu}}^{4+}$ to ${\mathrm{Pu}}^{3+}$. These species will tend to reduce Coulombic repulsion between ${\mathrm{H}}_{\mathrm{O}}$ and $V_{\mathrm{O}}$, resulting in faster diffusion. Our results provide important insights into the properties of hydrogen in plutonium oxides, which dictate the critical process of hydriding.

Low dimensional quantum materials have been extensively studied within the physics community, among which $\mathrm{C}{\mathrm{o}}^{2+}$ is one of the most popular magnetic ions to be embedded into the structural motif. $\mathrm{CoN}{\mathrm{b}}_2{\mathrm{O}}_6$ and $\mathrm{CoT}{\mathrm{a}}_2{\mathrm{O}}_6$ both have adopted an effective spin-1/2 state, with the former attracting enduring interest while the latter being rarely investigated. This difference can be attributed to their distinct crystal structures, not only in crystal systems, but more important, how the $\mathrm{Co}{\mathrm{O}}_6$ octahedra are connected. We successfully merge these two compounds together and map out the doping limits. We find out Ta and Nb dopants modify the properties of the parent compounds in subtle but different ways and construct a magnetic phase diagram of the $\mathrm{Co}{\left(\mathrm{N}{\mathrm{b}}_x\mathrm{T}{\mathrm{a}}_{1-x}\right)}_2{\mathrm{O}}_6$ $\left(0\le x\le 1\right)$ series. In addition, our results suggest that Ta-doping may pave a way to strengthen the lower-dimensional quantum phenomena in $\mathrm{CoN}{\mathrm{b}}_2{\mathrm{O}}_6$ before its crystal structure starts to melt down.

We present a machine-learning-guided first-principles study of rare-earth-free magnetic materials in the Fe-Co-S ternary system. A crystal-graph convolutional neural network (CGCNN) fine-tuned through three successive generations of iterative retraining, combined with adaptive genetic algorithm (AGA) structure searches, rapidly screens more than 300 000 substitutional structures and down-selects approximately 1400 candidates for density-functional-theory (DFT) validation. Spin-polarized DFT screening followed by phonon calculations identifies two dynamically stable tetragonal candidates, ${\mathrm{Fe}}_{17}{\mathrm{Co}}_4{\mathrm{S}}_4$ ($I4/m$) and ${\mathrm{Fe}}_2{\mathrm{Co}}_9{\mathrm{S}}_2$ ($I\overline{4}m2$), with saturation polarization $J_s\ge 1\mathrm{T}$ and energies approximately 0.10 eV/atom above the convex hull. These compounds emerge as the leading synthesis targets. Notably, ${\mathrm{Fe}}_{17}{\mathrm{Co}}_4{\mathrm{S}}_4$ exhibits $E_{\mathrm{hull}}=0.107$ eV/atom, $J_s=1.88\mathrm{T}$, and magnetocrystalline anisotropy $K_1=0.71 {\mathrm{MJ}/\mathrm{m}}^3$. We also identify eight near-hull phases ($E_{\mathrm{hull}}<0.01$ eV/atom) and independently rediscover the electrochemically prominent spinel ${\mathrm{FeCo}}_2{\mathrm{S}}_4$, providing an internal validation of the search strategy. Compared with our prior Fe-Co-C/Si/P studies, Fe-Co-S exhibits distinct stability and magnetization regimes consistent with the larger sulfur anion. The combination of machine learning with quantum simulations narrows synthesis targets for Fe-rich sulfides with high $J_s$ and moderate $K_1$, which are competitive with hard ferrites.