-

Shape memory ceramics are a unique family of shape memory materials with a wide variety of applications, such as ultra-high energy dissipation and high-temperature actuation. Along with significant progress in the experimental study of zirconia-based shape memory ceramics in recent years, computational simulations have exhibited powerful capabilities in revealing nano/microstructure-dependent deformation and failure mechanisms in these materials. In this work, we review the recent progress in understanding the phase transformation behavior in shape memory ceramics, focusing on computational modeling of zirconia-based ceramics. Electronic structure calculations have provided new data on the phase stability and surface properties of shape memory ceramics. At the nanometer length scale, molecular dynamics simulations have captured fundamental information about martensitic phase transformation and the effects of grain boundaries and defects on mechanical response of bi- and polycrystalline zirconia-based ceramics. At the micrometer length scale, advanced phase-field models have shown the ability to predict morphological evolution of microstructures and the corresponding mechanical responses in good agreement with experimental observations. Despite the recent multiscale computational advancements, further developments are required to establish processing-structure-property-performance relations that will lead to reliable and practical strategies for designing zirconia-based (or other) ceramics with improved and sustained shape memory responses. This article critically reviews current computational modeling techniques and provides an outlook for the study and design of the next generation of shape memory ceramics. Graphical abstract Download : Download high-res image (234KB) Download : Download full-size image

Electronic excitations and their dynamics are oftentimes at the foundation of how we use and probe materials. While recent experimental advances allow us to do so with unprecedented accuracy and time resolution, their interpretation relies on solid theoretical understanding. This can be provided by cutting-edge, first-principles theoretical-spectroscopy based on many-body perturbation theory (MBPT) and time-dependent density functional theory (TDDFT). In this work we review some of our recent results as successful examples for how electronic-structure methods lead to interesting insight into electronic excitations and deep understanding of modern materials. In many cases these techniques are accurate and even predictive, yet they rely on approximations to be computationally feasible. We illustrate the need for further theoretical understanding, using dielectric screening as an example in MBPT and faster, more accurate numerical integrators as a challenge for real-time TDDFT. Finally, we describe how incorporating online databases into computational materials research on excited electronic states can side-step the problem of high computational cost to facilitate materials design. Graphical abstract Download : Download high-res image (198KB) Download : Download full-size image

Charged impurities influence functional properties of two-dimensional materials and a detailed theoretical understanding of charged defects is required to enable a rational design of defect-engineered nanomaterials for applications in ultrathin devices. To achieve this goal, we have developed multiscale approaches that combine atomistic first-principles theories, such as density-functional theory, with coarse-grained continuum models, such as effective mass models. This allows us to model large supercells which are required to accurately describe the slow decay of the screened defect potential and the defect-induced changes in the electronic properties of the two-dimensional host material. I will describe the results of our multiscale calculations for charged defects in doped graphene and in transition-metal dichalcogenide monolayers which have revealed novel mechanisms for controlling and tuning the electronic structure of two-dimensional materials. Graphical abstract Download : Download high-res image (240KB) Download : Download full-size image

Computational material science has experienced three phases, i.e., structure-property calculations, crystal structure prediction, statistically driven design. In any case, structure-properties relation is the main outcome of any simulations and thus serves as a foundation for rational design, which requires a seamless collaboration among theoretical simulations, experimental characterizations and performance evaluations so as to accelerate the research and development (R&D) of functional materials. Rational design becomes feasible recently due to remarkable developments of computational algorithms and experimental characterizations. However, the realization of this concept remains challenges. In this review, we demonstrate how to design catalyst materials from scratch to their applications in sustainable energy storage and conversion. We conclude by pointing out several important challenges for future material design. Graphical abstract Schematics of rational material design which seamlessly combines experiments and theoretical simulations. Download : Download high-res image (111KB) Download : Download full-size image

Hematite ( α -Fe2O3) is one of the most widely studied materials for the application of water oxidation in the field of renewable energy. In order to push hematite toward maximal efficiency, a better understanding of the intrinsic limitations is essential. In this review, we cover our latest results on identifying and finding ways to overcome bottlenecks by modelling the electrochemical processes in hematite at three regimes: 1. at the bulk, the calculated hole effective mass is unfavorably high and conductivity can be improved through doping and alloying, 2. at the surface, catalysis is slow and involves long-lived surface states, but the catalytic overpotential can be reduced by adding a doping gradient or overlayers without strain, and 3. at the back contact the losses can be improved by selecting a proper metal contact. We provide unique analysis approaches that reveal fundamental understanding of electrochemistry with hematite. These models have made an impact on understanding experimental observations and have influenced subsequent theoretical investigations. Graphical abstract Download : Download high-res image (180KB) Download : Download full-size image

Microstructure sensitive design has a critical impact on the performance of engineering materials. The safety and performance requirements of critical components, as well as the cost of material and machining of Titanium components, make dovetailing of the microstructure imperative. This paper addresses the optimization of several microstructure design problems for Titanium components under specific design constraints using a feedback-aware data-driven solution methodology. In this study, the microstructure is modeled with an orientation distribution function (ODF) that measures the volumes of different crystallographic orientations. Two algorithms are used to sample the entire microstructure space followed by machine learning-aided identification of a minimal subset of ODF dimensions which is subsequently explored by targeted sampling. Conventional optimization methods lead to a unique microstructure rather than yielding a comprehensive space of optimal or near-optimal microstructures. Multiple solutions are crucial for the deployment of materials design for manufacturing as traditional manufacturing processes can only generate a limited set of microstructures. Our data sampling-based methodology not only outperforms or is on par with other optimization techniques in terms of the optimal property value, but also provides numerous near-optimal solutions, 3–4 orders of magnitude more than previous methods. Consequently, the proposed framework delivers a spectrum of optimal solutions in the microstructure space which can accelerate materials development and reduce manufacturing costs. Graphical abstract Flow diagram of the proposed methodology. There are two phases in the sampling process. In the first phase, two data generation algorithms are explored for dataset creation, namely partition and allocation. In the first step, we execute our data-generation algorithms (depicted by orange arrows) to generate many valid solutions for each set of constraint. In the second phase, a machine learning-guided feedback-aware sampling approach (depicted by green allows) extracts a combination of non- zero ODF dimensions. ODFs generated with these subset of combinations lead to optimal or near-optimal solutions. Download : Download high-res image (93KB) Download : Download full-size image

Highlights • StructOpt is an atomic optimization software package that uses genetic algorithms. • StructOpt determines atomic structures using multiobjective optimization. • StructOpt focuses on electron microscopy experimental data and simulated energies. • StructOpt can integrate already-existing experimental data simulations. • StructOpt’s code is open source and available on GitHub. StructOpt, an open-source structure optimization suite, applies genetic algorithm and particle swarm methods to obtain atomic structures that minimize an objective function. The objective function typically consists of the energy and the error between simulated and experimental data, which is typically applied to determine structures that minimize energy to the extent possible while also being fully consistent with available experimental data. We present example use cases including the structure of a metastable Pt nanoparticle determined from energetic and scanning transmission electron microscopy data, and the structure of an amorphous-nanocrystal composite determined from energetic and fluctuation electron microscopy data. StructOpt is modular in its construction and therefore is naturally extensible to include new materials simulation modules or new optimization methods, either written by the user or existing in other code packages. It uses the Message Passing Interface’s (MPI) dynamic process management functionality to allocate resources to computationally expensive codes on the fly, enabling StructOpt to take full advantage of the parallelization tools available in many scientific packages. Graphical abstract Download : Download high-res image (86KB) Download : Download full-size image

Highlights • Topological material is one of the hot materials in condensed matter physics. • However, less research has been done on the properties of defects for them. • Our study helps to gain insight into the properties of them. • We use the DFT method to study the this subject. • We find that the topological properties of NbAs are relatively stable. Niobium Arsenic (NbAs), presenting some novel properties, is a high-profile topological material in condensed matter physics. Our study focuses on the NbAs after hydrogen (H) and helium (He) radiation. There are a variety of defects induced. We investigate into the electronic properties of defects by analyzing electronic dispersion curves and density of states based on the DFT method. Meanwhile the formation energy of these defects are calculated. The impacts of radiation are various in different defects. Defects of interstitial H and He atom do not affect the topological properties of NbAs at low defect concentration. However, the other defects destruct part of the Weyl points, especially in the defects of vacancy and substitution As for Nb. In addition, Fermi levels of these defects shift and the degenerate bands are separate. From the perspective of formation energy, the defect of H interstitial is the easiest one to be formed in all the defects.

Highlights • ANN and GAs are used to comprehend material behavior under extreme conditions. • Properties of modern construction materials at elevated temperatures are derived. • AI could modernize materials science and structural engineering for harsh conditions. The construction industry has adopted a number of new building materials over the past few years. While these materials are specifically designed to achieve improved strength and durability characteristics at ambient conditions, the performance of modern construction materials (MCMs) under extreme conditions such as fire is still not understood. Under elevated temperatures, MCMs not only undergo a series of physio-chemical degradations, but these degradations are often of a much severe magnitude than that in traditional construction materials (TCMs). Despite ongoing efforts, there continues to be a lack of guidance/provisions on how to account for such temperature-induced degradations in MCMs. This adds another dimension of complexity to researchers and engineers seeking to carry out fire resistance evaluation and also presents a major challenge towards promoting standardization and performance-based solutions for fire engineering applications. In order to bridge this knowledge gap, this paper presents a methodology to develop temperature-dependent material models for MCMs such as high strength/performance concrete (HSC/HPC), high/very high strength steels (HSS/VHSS), and fiber-reinforced polymer (FRP) composites, using two techniques of artificial intelligence (AI) namely: artificial neural networks (ANNs) and evolutionary genetic algorithms (GAs). The outcome of this study showcases the merit of integrating AI into understanding the complex behavior of MCMs under fire conditions as well as in deriving temperature-dependent material models for these materials. Graphical abstract Download : Download high-res image (61KB) Download : Download full-size image