Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userCrystal structure prediction via combining graph network and Bayesian optimization
102
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
We developed a density functional theory-free approach for crystal structure prediction via combing graph network (GN) and Bayesian optimization (BO). GN is adopted to establish the correlation model between crystal structure and formation enthalpies. BO is to accelerate searching crystal structure with optimal formation enthalpy. The approach of combining GN and BO for crystal Structure Searching (GN-BOSS), in principle, can predict crystal structure at given chemical compositions without additional constraints on cell shapes and lattice symmetries. The applicability and efficiency of GN-BOSS approach is then verified via solving the classical Ph-vV challenge. It can correctly predict the crystal structures of 24 binary compounds from scratch with averaged computational cost ~ 30 minutes each by only one CPU core. GN-BOSS approach may open a new avenue to data-driven crystal structural prediction without using the expensive DFT calculations.
rate research
Read More
Crystal structure prediction is now playing an increasingly important role in discovery of new materials. Global optimization methods such as genetic algorithms (GA) and particle swarm optimization (PSO) have been combined with first principle free energy calculations to predict crystal structures given composition or only a chemical system. While these approaches can exploit certain crystal patterns such as symmetry and periodicity in their search process, they usually do not exploit the large amount of implicit rules and constraints of atom configurations embodied in the large number of known crystal structures. They currently can only handle crystal structure prediction of relatively small systems. Inspired by the knowledge-rich protein structure prediction approach, herein we explore whether known geometric constraints such as the atomic contact map of a target crystal material can help predict its structure given its space group information. We propose a global optimization based algorithm, CMCrystal, for crystal structure reconstruction based on atomic contact maps. Based on extensive experiments using six global optimization algorithms, we show that it is viable to reconstruct the crystal structure given the atomic contact map for some crystal materials but more constraints are needed for other target materials to achieve successful reconstruction. This implies that atomic interaction information learned from existing materials can be used to improve crystal structure prediction.
Structure is the most basic and important property of crystalline solids; it determines directly or indirectly most materials characteristics. However, predicting crystal structure of solids remains a formidable and not fully solved problem. Standard theoretical tools for this task are computationally expensive and at times inaccurate. Here we present an alternative approach utilizing machine learning for crystal structure prediction. We developed a tool called Crystal Structure Prediction Network (CRYSPNet) that can predict the Bravais lattice, space group, and lattice parameters of an inorganic material based only on its chemical composition. CRYSPNet consists of a series of neural network models, using as inputs predictors aggregating the properties of the elements constituting the compound. It was trained and validated on more than 100,000 entries from the Inorganic Crystal Structure Database. The tool demonstrates robust predictive capability and outperforms alternative strategies by a large margin. Made available to the public (at https://github.com/AuroraLHT/cryspnet), it can be used both as an independent prediction engine or as a method to generate candidate structures for further computational and/or experimental validation.
Crystal structure prediction is a central problem of theoretical crystallography and materials science, which until mid-2000s was considered intractable. Several methods, based on either energy landscape exploration$^{1,2}$ or, more commonly, global optimization$^{3-8}$, largely solved this problem and enabled fully non-empirical computational materials discovery$^{9,10}$. A major shortcoming is that, to avoid expensive calculations of the entropy, crystal structure prediction was done at zero Kelvin and searched for the global minimum of the enthalpy, rather than free energy. As a consequence, high-temperature phases (especially those which are not quenchable to zero temperature) could be missed. Here we develop an accurate and affordable solution, enabling crystal structure prediction at finite temperatures. Structure relaxation and fully anharmonic free energy calculations are done by molecular dynamics with a force field (which can be anything from a parametric force field for simpler cases to a trained on-the-fly machine learning interatomic potential), the errors of which are corrected using thermodynamic perturbation theory to yield accurate ab initio results. We test the accuracy of this method on metals (probing the P-T phase diagram of Al and Fe), a refractory intermetallide (WB), and a significantly ionic ceramic compound (Earth-forming silicate MgSiO3 at pressures and temperatures of the Earths lower mantle). We find that the hcp-phase of aluminum has a wider stability field than previously thought, and the temperature-induced transition $alpha$-$beta$ in WB occurs at 2789 K. It is also found that iron has hcp structure at conditions of the Earths inner core, and the much debated (and important for constraining Earths thermal structure) Clapeyron slope of the post-perovskite phase transition in MgSiO3 is 5.88 MPa/K.
Machine learning (ML) models utilizing structure-based features provide an efficient means for accurate property predictions across diverse chemical spaces. However, obtaining equilibrium crystal structures typically requires expensive density functional theory (DFT) calculations, which limits ML-based exploration to either known crystals or a small number of hypothetical crystals. Here, we demonstrate that the application of Bayesian optimization with symmetry constraints using a graph deep learning energy model can be used to perform DFT-free relaxations of crystal structures. Using this approach to significantly improve the accuracy of ML-predicted formation energies and elastic moduli of hypothetical crystals, two novel ultra-incompressible hard materials MoWC2 (P63/mmc) and ReWB (Pca21) were identified and successfully synthesized via in-situ reactive spark plasma sintering from a screening of 399,960 transition metal borides and carbides. This work addresses a critical bottleneck to accurate property predictions for hypothetical materials, paving the way to ML-accelerated discovery of new materials with exceptional properties.
Prediction of stable crystal structures at given pressure-temperature conditions, based only on the knowledge of the chemical composition, is a central problem of condensed matter physics. This extremely challenging problem is often termed crystal structure prediction problem, and recently developed evolutionary algorithm USPEX (Universal Structure Predictor: Evolutionary Xtallography) made an important progress in solving it, enabling efficient and reliable prediction of structures with up to ~40 atoms in the unit cell using ab initio methods. Here we review this methodology, as well as recent progress in analyzing energy landscape of solids (which also helps to analyze results of USPEX runs). We show several recent applications - (1) prediction of new high-pressure phases of CaCO3, (2) search for the structure of the polymeric phase of CO2 (phase V), (3) high-pressure phases of oxygen, (4) exploration of possible stable compounds in the Xe-C system at high pressures, (5) exotic high-pressure phases of elements boron and sodium.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
