Search

One of the most effective methods to enhance the strength of aluminum alloys involves modifying grain boundaries (GBs) through solute segregation. However, the fundamental mechanisms of solute segregation and their impacts on material properties remain elusive. In this study, we implemented highly efficient hybrid Monte Carlo and molecular dynamics (MCMD) algorithms in the graphics process units molecular dynamics (GPUMD) package. Using this efficient MCMD approach combined with a general-purpose machine-learning-based neuroevolution potential (NEP) for 16 elemental metals and their alloys, we simulated the segregation of 15 solutes in polycrystalline Al. Our results elucidate the segregation behavior and trends of 15 solutes in polycrystalline Al. Additionally, we investigated the impact of solutes on the strength of polycrystalline Al. The mechanisms underlying solute strengthening and embrittlement were analyzed at the atomistic level, revealing the importance of GB cohesion, as well as the nucleation and movement of Shockley dislocations, in determining the material's strength. We anticipate that our developed methods, along with our insights into solute segregation behavior in polycrystalline Al, will be valuable for the design of Al alloys and other multi-component materials, including medium-entropy materials, high-entropy materials, and complex concentrated alloys., Comment: 10 pages, 6 figures

Machine-learned potentials (MLPs) have exhibited remarkable accuracy, yet the lack of general-purpose MLPs for a broad spectrum of elements and their alloys limits their applicability. Here, we present a feasible approach for constructing a unified general-purpose MLP for numerous elements, demonstrated through a model (UNEP-v1) for 16 elemental metals and their alloys. To achieve a complete representation of the chemical space, we show, via principal component analysis and diverse test datasets, that employing one-component and two-component systems suffices. Our unified UNEP-v1 model exhibits superior performance across various physical properties compared to a widely used embedded-atom method potential, while maintaining remarkable efficiency. We demonstrate our approach's effectiveness through reproducing experimentally observed chemical order and stable phases, and large-scale simulations of plasticity and primary radiation damage in MoTaVW alloys. This work represents a significant leap towards a unified general-purpose MLP encompassing the periodic table, with profound implications for materials science., Comment: Main text with 17 pages and 8 figures; supplementary with 26 figures and 4 tables; source code and training/test data available

Predicting high temperature superconductors has long been a great challenge. A major difficulty is how to predict the transition temperature Tc of superconductors. Recently, progress in material informatics has led to a number of machine learning models predicting Tc, which greatly improves the efficiency of prediction. Unfortunately, prevailing models have not shown adequate physical rationality and generalization ability to find new high temperature superconductors, yet. In this work, in order to give a trustable prediction on the unexplored materials, we built a bond-sensitive graph neural network (BSGNN), which is optimized to process the information of chemical bond and electron interaction in the crystal lattice, to predict the Tc maximum of each type of superconducting materials. On the basis of the domain knowledge considered in the data preparation and algorithm design, our model revealed a relevance between the Tc-Tc maximum and chemical bonds. The results indicate that shorter bond length is favored by high Tc, which is in accordance with previous human experience. Moreover, it also shows that some specific chemical elements are favored by high Tc, which is beyond what human experts already knew. It gives a convenient guidance for searching high temperature superconductors in materials database, by ruling out the materials that could never have high Tc.

Simulating collision cascades and radiation damage poses a long-standing challenge for existing interatomic potentials, both in terms of accuracy and efficiency. Machine-learning based interatomic potentials have shown sufficiently high accuracy for radiation damage simulations, but most existing ones are still not efficient enough to model high-energy collision cascades with sufficiently large space and time scales. To this end, we here extend the highly efficient neuroevolution potential (NEP) framework by combining it with the Ziegler-Biersack-Littmark (ZBL) screened nuclear repulsion potential, obtaining a NEP-ZBL framework. We train a NEP-ZBL model for tungsten and demonstrate its accuracy in terms of the elastic properties, melting point, and various energetics of defects that are relevant for radiation damage. We then perform large-scale molecular dynamics simulations with the NEP-ZBL model with up to 8.1 million atoms and 240 ps (using a single 40-GB A100 GPU) to study the difference of primary radiation damage in both bulk and thin-foil tungsten. While our findings for bulk tungsten are consistent with existing results simulated by embedded atom method (EAM) models, the radiation damage differs significantly in foils and shows that larger and more vacancy clusters as well as smaller and fewer interstitial clusters are produced due to the presence of a free surface.

Recently a novel two-dimensional (2D) C$_{60}$ based crystal called quasi-hexagonal-phase fullerene (QHPF) has been fabricated and demonstrated to be a promising candidate for 2D electronic devices [Hou et al. Nature 606, 507-510 (2022)]. We construct an accurate and transferable machine-learned potential to study heat transport and related properties of this material, with a comparison to the face-centered-cubic bulk-phase fullerene (BPF). Using the homogeneous nonequilibrium molecular dynamics and the related spectral decomposition methods, we show that the thermal conductivity in QHPF is anisotropic, which is 137(7) W/mK at 300 K in the direction parallel to the cycloaddition bonds and 102(3) W/mK in the perpendicular in-plane direction. By contrast, the thermal conductivity in BPF is isotropic and is only 0.45(5) W/mK. We show that the inter-molecular covalent bonding in QHPF plays a crucial role in enhancing the thermal conductivity in QHPF as compared to that in BPF. The heat transport properties as characterized in this work will be useful for the application of QHPF as novel 2D electronic devices., Comment: 11 pages, 12 figures

In a previous paper [Physical Review B \textbf{103}, 035417 (2021)], we showed that the equilibrium molecular dynamics (EMD) method can be used to compute the apparent thermal conductivity of finite systems. It has been shown that the apparent thermal conductivity from EMD for a system with domain length $2L$ is equal to that from nonequilibrium molecular dynamics (NEMD) for a system with domain length $L$. Taking monolayer silicence with an accurate machine learning potential as an example, here we show that the thermal conductivity values from EMD and NEMD agree for the same domain length if the NEMD is applied with periodic boundary conditions in the transport direction. Our results thus establish an exact equivalence between EMD and NEMD for thermal conductivity calculations., Comment: 5 pages, 5 figures

We present our latest advancements of machine-learned potentials (MLPs) based on the neuroevolution potential (NEP) framework introduced in [Fan et al., Phys. Rev. B 104, 104309 (2021)] and their implementation in the open-source package GPUMD. We increase the accuracy of NEP models both by improving the radial functions in the atomic-environment descriptor using a linear combination of Chebyshev basis functions and by extending the angular descriptor with some four-body and five-body contributions as in the atomic cluster expansion approach. We also detail our efficient implementation of the NEP approach in graphics processing units as well as our workflow for the construction of NEP models, and we demonstrate their application in large-scale atomistic simulations. By comparing to state-of-the-art MLPs, we show that the NEP approach not only achieves above-average accuracy but also is far more computationally efficient. These results demonstrate that the GPUMD package is a promising tool for solving challenging problems requiring highly accurate, large-scale atomistic simulations. To enable the construction of MLPs using a minimal training set, we propose an active-learning scheme based on the latent space of a pre-trained NEP model. Finally, we introduce three separate Python packages, GPYUMD, CALORINE, and PYNEP, which enable the integration of GPUMD into Python workflows., Comment: 29 pages, 15 figures, code and data available

We study the interfacial thermal conductance of grain boundaries (GBs) between monolayer graphene and hexagonal boron nitride (h-BN) sheets using a combined atomistic approach. First, realistic samples containing graphene/h-BN GBs with different tilt angles are generated using the phase-field crystal (PFC) model developed recently [P. Hirvonen \textit{et al.}, Phys. Rev. B \textbf{100}, 165412 (2019)] that captures slow diffusive relaxation inaccessible to molecular dynamics (MD) simulations. Then, large-scale MD simulations using the efficient GPUMD package are performed to assess heat transport and rectification properties across the GBs. We find that lattice mismatch between the graphene and h-BN sheets plays a less important role in determining the interfacial thermal conductance as compared to the tilt angle. In addition, we find no significant thermal rectification effects for these GBs., Comment: 8 pages, 8 figures

We propose a way to properly interpret the apparent thermal conductivity obtained for finite systems using equilibrium molecular dynamics simulations (EMD) with fixed or open boundary conditions in the transport direction. In such systems the heat current autocorrelation function develops negative values after a correlation time which is proportional to the length of the simulation cell in the transport direction. Accordingly, the running thermal conductivity develops a maximum value at the same correlation time and eventually decays to zero. By comparing EMD with nonequilibrium molecular dynamics (NEMD) simulations, we conclude that the maximum thermal conductivity from EMD in a system with domain length 2L is equal to the thermal conductivity from NEMD in a system with domain length L. This facilitates the use of nonperiodic-boundary EMD for thermal transport in finite samples in close correspondence to NEMD., Comment: 7pages, 8 figures

Single-walled carbon nanotubes (SWCNTs) in their pristine form have high thermal conductivity whose further improvement has attracted a lot of interest. Some theoretical studies have suggested that the thermal conductivity of a $(10,10)$ SWCNT is dramatically enhanced by C$_{60}$ fullerene encapsulation. However, recent experiments on SWCNT bundles show that fullerene encapsulation leads to a reduction rather than an increase in thermal conductivity. Here, we employ three different molecular dynamics methods to study the influence of C$_{60}$ encapsulation on heat transport in a $(10,10)$ SWCNT. All the three methods consistently predict a reduction of the thermal conductivity of $(10,10)$ SWCNT upon C$_{60}$ encapsulation by $20\%-30\%$, in agreement with experimental results on bundles of SWCNTs. We demonstrate that there is a simulation artifact in the Green-Kubo method which gives anomalously large thermal conductivity from artificial convection. Our results show that the C$_{60}$ molecules conduct little heat compared to the outer SWCNT and reduce the phonon mean free paths of the SWCNT by inducing extra phonon scattering. We also find that the thermal conductivity of a $(10,10)$ SWCNT monotonically decreases with increasing filling ratio of C$_{60}$ molecules., Comment: 26 pages, 10 figures

Silicon is an important material and many empirical interatomic potentials have been developed for atomistic simulations of it. Among them, the Tersoff potential and its variants are the most popular ones. However, all the existing Tersoff-like potentials fail to reproduce the experimentally measured thermal conductivity of diamond silicon. Here we propose a modified Tersoff potential and develop an efficient open source code called GPUGA (graphics processing units genetic algorithm) based on the genetic algorithm and use it to fit the potential parameters against energy, virial and force data from quantum density functional theory calculations. This potential, which is implemented in the efficient open source GPUMD (graphics processing units molecular dynamics) code, gives significantly improved descriptions of the thermal conductivity and phonon dispersion of diamond silicon as compared to previous Tersoff potentials and at the same time well reproduces the elastic constants. Furthermore, we find that quantum effects on the thermal conductivity of diamond silicon at room temperature are non-negligible but small: using classical statistics underestimates the thermal conductivity by about 10\% as compared to using quantum statistics., Comment: 9 pages, 6 figures

Recent progress in material data mining has been driven by high-capacity models trained on large datasets. However, collecting experimental data (real data) has been extremely costly since the amount of human effort and expertise required. Here, we develop a novel transfer learning strategy to address small or insufficient data problem. This strategy realizes the fusion of real and simulated data, and the augmentation of training data in data mining procedure. For a specific task of image segmentation, this strategy can generate synthetic images by fusing physical mechanism of simulated images and "image style" of real images. The result shows that the model trained with the acquired synthetic images and 35% of the real images outperforms the model trained on all real images. As the time required to generate synthetic data is almost negligible, this strategy is able to reduce the time cost of real data preparation by roughly 65%., Comment: 17 pages, technical report