Machine learning, phase stability, and disorder with the Automatic Flow Framework for Materials Discovery

Abstract

Traditional materials discovery approaches - relying primarily on laborious experiments - have controlled the pace of technology. Instead, computational approaches offer an accelerated path: high-throughput exploration and characterization of virtual structures. These ventures, performed by automated ab-initio frameworks, have rapidly expanded the volume of programmatically-accessible data, cultivating opportunities for data-driven approaches. Herein, a collection of robust characterization methods are presented, implemented within the Automatic Flow Framework for Materials Discovery (AFLOW), that leverages materials data for the prediction of phase diagrams and properties of disordered materials. These methods directly address the issue of materials synthesizability, bridging the gap between simulation and experiment. Powering these predictions is the AFLOW.org repository for inorganic crystals, the largest and most comprehensive database of its kind, containing more than 2 million compounds with about 100 different properties computed for each. As calculated with standardized parameter sets, the wealth of data also presents a favorable learning environment. Machine learning algorithms are employed for property prediction, descriptor development, design rule discovery, and the identification of candidate functional materials. When combined with physical models and intelligently formulated descriptors, the data becomes a powerful tool, facilitating the discovery of new materials for applications ranging from high-temperature superconductors to thermoelectrics. These methods have been validated by the synthesis of two new permanent magnets introduced herein - the first discovered by computational approaches.

arXiv commentPhD thesis, 519 pages, 135 figures

References in corpus (77)

• Scikit-learn: Machine Learning in Python
• Emergence of scaling in random networks
• SMOTE: Synthetic Minority Over-sampling Technique
• Topological Insulators
• The structure and function of complex networks
• Community structure in social and biological networks
• The SIESTA method for ab initio order-N materials simulation
• Efficient index handling of multidimensional periodic boundary conditions
• Necessary and Sufficient Elastic Stability Conditions in Various Crystal Systems
• Intrinsic Correlation between Hardness and Elasticity in Polycrystalline Materials and Bulk Metallic Glasses
• ANI-1: An extensible neural network potential with DFT accuracy at force field computational cost
• Two-dimensional materials from high-throughput computational exfoliation of experimentally known compounds
• A General-Purpose Machine Learning Framework for Predicting Properties of Inorganic Materials
• AFLOW: An automatic framework for high-throughput materials discovery
• High-temperature superconductivity in iron-based materials
• Big Data of Materials Science - Critical Role of the Descriptor
• The virtual crystal approximation revisited: Application to dielectric and piezoelectric properties of perovskites
• Experimental demonstration of the topological surface states protected by the time-reversal symmetry
• A new platform for topological quantum phenomena : Topological Insulator states in thermoelectric Heusler-related ternary compounds
• Lattice dynamics of anharmonic solids from first principles
• Universal Fragment Descriptors for Predicting Electronic Properties of Inorganic Crystals
• Self-consistent phonon calculations of lattice dynamical properties in cubic SrTiO${}_3$ with first-principles anharmonic force constants
• Superfluidity in topologically nontrivial flat bands
• AiiDA: Automated Interactive Infrastructure and Database for Computational Science
• Realization of spin gapless semiconductors: the Heusler compound Mn2CoAl
• High-temperature surface superconductivity in topological flat-band systems
• How to represent crystal structures for machine learning: towards fast prediction of electronic properties
• Discovery of low thermal conductivity compounds with first-principles anharmonic lattice dynamics calculations and Bayesian optimization
• Temperature dependent effective third order interatomic force constants from first principles
• How to calculate the degree of spin polarization in ferromagnets?
• Generating derivative structures: Algorithm and applications
• Predicting Crystal Structures with Data Mining of Quantum Calculations
• Entropy driven stabilization of energetically unstable crystal structures explained from first principles theory
• Machine learning with systematic density-functional theory calculations: Application to melting temperatures of single and binary component solids
• Materials Cartography: Representing and Mining Material Space Using Structural and Electronic Fingerprints
• First-Principles Theory of Anharmonicity and the Inverse Isotope Effect in Superconducting Palladium-Hydride Compounds
• Half-Heusler semiconductors as piezoelectrics
• Predicting the thermodynamic stability of perovskite oxides using machine learning models
• Discovery of a superhard iron tetraboride superconductor
• Large zero-field cooled exchange-bias in bulk Mn2PtGa
• Accuracy of generalized gradient approximation functionals for density functional perturbation theory calculations
• New Superconducting and Semiconducting Fe-B Compounds Predicted with an Ab Initio Evolutionary Search
• Hole-Doped Cuprate High Temperature Superconductors
• Thermal Expansion and Gruneisen Parameters of Amorphous Silicon: A Realistic Model Calculation
• Phase stability and structural temperature dependence in sodium niobate: A high resolution powder neutron diffraction study
• Spectral descriptors for bulk metallic glasses based on the thermodynamics of competing crystalline phases
• Quasiparticle Effects in the Bulk and Surface-State Bands of Bi2Se3 and Bi2Te3 Topological Insulators
• Large Noncollinearity and Spin Reorientation in the Novel Mn2RhSn Heusler Magnet
• Magnetic and transport properties of Mn2CoAl oriented thin films
• Design and discovery of a novel Half-Heusler transparent hole conductor made of all-metallic heavy elements
• Glass-like thermal conductivity in SrTiO3 thermoelectrics induced by A-site vacancies
• Structure prediction based on ab initio simulated annealing for boron nitride
• Exchange interactions and Curie temperatures in Mn2CoZ compounds
• Theoretical study of metal borides stability
• First-principles DFT+GW study of oxygen vacancies in rutile TiO2
• GW quasiparticle bandgaps of anatase TiO2 starting from DFT+U
• Anomalous thermal conductivity and suppression of negative thermal expansion in ScF3
• Prediction of new crystal structure phases in metal borides: a lithium monoboride analog to MgB2
• Effective and accurate representation of extended Bloch states on finite Hilbert spaces
• Dynamic of a non homogeneously coarse grained system
• Finding the stable structures of W${}_x$N${}_1$${}_{-}$${}_x$ with an $ab$$-$$initio$ high-throughput approach
• Itinerant and local magnetic moments in ferrimagnetic Mn2CoGa thin films probed by x-ray magnetic linear dichroism: experiment and ab initio theory
• High-Throughput Screening of Perovskite Alloys for Piezoelectric Performance and Formability
• Looking for Design in Materials Design
• Structure maps for hcp metals from first principles calculations
• Exceptionally strong magnetism in 4d perovskites RTcO3 (R=Ca,Sr,Ba)
• Phonon heat conduction in layered anisotropic crystals
• Specific heat and heat conductivity of the BaTiO3 polycrystalline films with the thickness in the range 20 - 1100 nm
• Pressure-dependence of electron-phonon coupling and the superconducting phase in hcp Fe - a linear response study
• Calculation of solubility in titanium alloys from first-principles
• Surface-electronic structure of La(0001) and Lu(0001)
• Discovery of several large families of Topological Insulator classes with backscattering-suppressed spin-polarized single-Dirac-cone on the surface
• The molybdenum-titanium phase diagram evaluated from ab-initio calculations
• Stable ordered structures of binary technetium alloys from first principles
• Anisotropic superconductivity mediated by phonons in layered compounds with weak screening effect
• Electronic structure and self energies of randomly substituted solids using density functional theory and model calculations
• An Innovative Approach to Balancing Chemical-Reaction Equations: A Simplified Matrix-Inversion Technique for Determining The Matrix Null Space