Mixing Guide Elements | Natural Computational Science

Composite materials (CCMs), which are formed by mixing multiple elements, are undoubtedly of great interest in materials science and engineering. The availability of a large number of mixed elements in the material lattice greatly expands the chemical space for materials design. Accordingly, this extension will help materials scientists access previously unexplored physical domains, resulting in improved and exotic material properties and enabling researchers to explore unprecedented applications. will be

In this issue Natural materials, which focuses on recent developments in the burgeoning field of complex element bonding, summarizes expert views on CCM design opportunities for extending material capabilities.In particular we natural computational science We present a collection of manuscripts that provide insight into key issues in developing computational methods for guiding elemental mixing in CCM.

Needless to say, the field of computational materials science has experienced tremendous growth over the past decades when it comes to steering materials design. For example, first-principles methods at atomic and electronic scales have enabled the computation of defect energetics and electronic structure, providing insightful knowledge for dopant design in semiconductor applications such as transistors and photovoltaics. rice field.¹Another example is mesoscale calculation methods such as the phase-field method.², has been extensively used to investigate material properties related to kinetics and microstructural evolution. However, these conventional techniques are limited when dealing with new complications arising in CCM, such as low crystal symmetry, increased number of competing phases, and imbalance of interactions between various pairs of atoms. It may have no effect. Therefore, there is a demand in the computational science community to make effective use of available computational power to guide the design of CCMs.

Physics-based models continue to provide essential advances in the materials design process, but face no obstacles. A central computational challenge is to accurately model the complexity while keeping the computation affordable. For example, one of the key tasks in materials engineering is to predict the defect properties of materials and further relate these properties to application design. When modeling defects in CCM such as vacancies and interstitial atoms, it is computationally expensive to sample and calculate the increasing number of nonuniform defect sites in different local chemical environments. In a Perspective, Xie Zhang and colleagues outline the challenges and opportunities for extending traditional defect physics models to his CCM. They argue that the traditional defect energy model can be extended to CCM by defining effective formation energies using advanced computational tools such as statistical methods, sampling techniques and configuration generation tools. Zhang et al. further point out that computing the energy of such a vast configuration space using purely quantum mechanical methods such as density functional theory (DFT) is very costly. A level of precision is required.

To address some of the challenges faced by traditional physics-based models, artificial intelligence (AI)-based techniques for learning models from data have been identified as a promising place in this field. For example, in a widely used classical potential, as introduced in his Review by Alberto Ferrari and colleagues, the ubiquitous and technically relevant short-range order (SRO) in CCM—the local element A kind of order—it has become impossible to describe precisely. To reliably mimic high-dimensional interatomic relationships, a vast number of parameters are required for different mixed elements. Ferrari and his colleagues argue that machine learning interatomic potentials such as low-rank potentials and moment tensor potentials can be used to effectively investigate his SRO in CCM. Similarly, Dierk Raabe and colleagues noted in a Perspective that his AI model considers more constraints during the optimization process of element selection and compositional design to provide better guidance on controlling impurities during the manufacturing process. looking into how to do it. As an example, constraints related to recyclability of mixed elements can be considered for more sustainable material production. The importance of trace impurity control in alloy design is also emphasized. Natural materials Q&A with Zhi-Wei Shan.

It is worth emphasizing that AI-based models can complement rather than simply replace physics-based models. On the one hand, AI approaches can greatly expand the search space to identify more complex physical relationships for the design of CCM, but this large dimensionality of the search space can affect efficiency. . Physics-based models, on the other hand, rely on established functional forms that can somewhat limit new discoveries, but can guide the design process more efficiently. Incorporating physical relationships such as the laws of thermodynamics and DFT predictions as constraints into AI models is a promising strategy in areas where optimization and search efficiency can be improved through a careful balance of exploration and exploitation. is.

This focus also features key research papers. Natural materials, exemplifies how exotic properties can be achieved through sophisticated engineering of element mixing. For example, an article by Hang Xue and colleagues presents an interstitial solute stabilization strategy that produces dense, highly stable and coherent nanoprecipitates in his Sc-doped Al–Cu–Mg–Ag alloy. , allowing Al alloys to reach unprecedented creep resistance. It also has exceptional tensile strength at high temperatures. In another article, Jinlong Du and colleagues report a reversible local disorder-to-order transition of precipitates through careful element design in multicomponent metallic alloys that enable high radiation resistance at high temperatures. doing. Finally, an article by Jiadong Zhou and colleagues reports the synthesis of various his 2D CCMs that achieve tunable material properties such as ferromagnetism and superconductivity. We believe similar success stories will continue to inspire the community, especially with guidance from computational and theoretical insights.

Finally, I would like to emphasize the fact that CCM design requires interdisciplinary collaboration between experimentalists, theorists, and computational scientists. Theoretical and computational insights help experimenters better navigate vast design spaces, but theoretical mechanisms also include validating findings and providing constructive feedback to experimenters. We need new experimental data toWe believe that this joint focus natural computational science and Natural materials inspires new collaborations that accelerate new discoveries in materials engineering.