A5073635313- Machine Learning in Materials Science
- X-ray Diffraction in Crystallography
- High Entropy Alloys Studies
- Magnetic properties of thin films
- Advanced Materials Characterization Techniques
- Hydrogen Storage and Materials
- Electron and X-Ray Spectroscopy Techniques
- Electronic and Structural Properties of Oxides
- Nuclear Materials and Properties
- Semiconductor materials and devices
- Catalytic Processes in Materials Science
- Electrocatalysts for Energy Conversion
- Magnetic Properties and Applications
- High-Temperature Coating Behaviors
- Corrosion Behavior and Inhibition
- Computational Drug Discovery Methods
- Metal and Thin Film Mechanics
- Magnetic and transport properties of perovskites and related materials
- Hydrogen embrittlement and corrosion behaviors in metals
- Multiferroics and related materials
- Electrochemical Analysis and Applications
- Ferroelectric and Negative Capacitance Devices
- Magnetic Properties of Alloys
- Ferroelectric and Piezoelectric Materials
- Phase-change materials and chalcogenides
Schwartz/Reisman Emergency Medicine Institute
2024-2025
Vector Institute
2023-2025
University of Toronto
2021-2024
University of New Brunswick
2022-2024
Structural Genomics Consortium
2024
Natural Resources Canada
2022-2024
National Institute of Standards and Technology
2010-2023
Material Measurement Laboratory
2011-2023
Materials Science & Engineering
2022-2023
University of Maryland, College Park
2001-2022
Coupling artificial intelligence with high-throughput experimentation accelerates discovery of amorphous alloys.
Active learning-the field of machine learning (ML) dedicated to optimal experiment design-has played a part in science as far back the 18th century when Laplace used it guide his discovery celestial mechanics. In this work, we focus closed-loop, active learning-driven autonomous system on another major challenge, advanced materials against exceedingly complex synthesis-processes-structure-property landscape. We demonstrate an methodology for functional inorganic compounds which allow...
The Materials Genome Initiative, a national effort to introduce new materials into the market faster and at lower cost, has made significant progress in computational simulation modeling of materials. To build on this progress, large amount experimental data for validating these models, informing more sophisticated ones, will be required. High-throughput experimentation generates volumes using combinatorial synthesis rapid measurement techniques, making it an ideal complement bring...
High throughput (combinatorial) materials science methodology is a relatively new research paradigm that offers the promise of rapid and efficient screening, optimization, discovery. The started in pharmaceutical industry but was rapidly adopted to accelerate wide variety areas. experiments are characterized by synthesis “library” sample contains variation interest (typically composition), localized measurement schemes result massive data sets. Because collected at same time on sample, they...
Traditional machine learning (ML) metrics overestimate model performance for materials discovery.
Abstract Recent advances in machine learning (ML) have led to substantial performance improvement material database benchmarks, but an excellent benchmark score may not imply good generalization performance. Here we show that ML models trained on Materials Project 2018 can severely degraded new compounds 2021 due the distribution shift. We discuss how foresee issue with a few simple tools. Firstly, uniform manifold approximation and projection (UMAP) be used investigate relation between...
Extensive efforts to gather materials data have largely overlooked potential redundancy. In this study, we present evidence of a significant degree redundancy across multiple large datasets for various material properties, by revealing that up 95% can be safely removed from machine learning training with little impact on in-distribution prediction performance. The redundant is related over-represented types and does not mitigate the severe performance degradation out-of-distribution samples....
Abstract Scientific machine learning (ML) aims to develop generalizable models, yet assessments of generalizability often rely on heuristics. Here, we demonstrate in the materials science setting that heuristic evaluations lead biased conclusions ML and benefits neural scaling, through out-of-distribution (OOD) tasks involving unseen chemistry or structural symmetries. Surprisingly, many good performance across including boosted trees. However, analysis representation space shows most test...
Chemical and structural heterogeneity the resulting interaction of coexisting phases can lead to extraordinary behaviours in oxides, as observed piezoelectric materials at morphotropic phase boundaries relaxor ferroelectrics. However, such phenomena are rare metallic alloys. Here we show that, by tuning presence textured Co(1-x)Fe(x) thin films, effective magnetostriction λ(eff) large 260 p.p.m. be achieved low-saturation field ~10 mT. Assuming λ(100) is dominant component, this number...
With their ability to rapidly elucidate composition-structure-property relationships, high-throughput experimental studies have revolutionized how materials are discovered, optimized, and commercialized. It is now possible synthesize characterize libraries that systematically address thousands of individual cuts fabrication parameter space. An unresolved issue remains transforming structural characterization data into phase mappings. This difficulty related the complex information present in...
Insufficient availability of molten salt corrosion-resistant alloys severely limits the fruition a variety promising technologies that could otherwise have significant societal impacts. To accelerate alloy development for applications and develop fundamental understanding corrosion in these environments, here an integrated approach is presented using set high-throughput (HTP) synthesis, testing, modeling coupled with automated characterization machine learning. By this approach, broad range...
Abstract Lack of rigorous reproducibility and validation are significant hurdles for scientific development across many fields. Materials science, in particular, encompasses a variety experimental theoretical approaches that require careful benchmarking. Leaderboard efforts have been developed previously to mitigate these issues. However, comprehensive comparison benchmarking on an integrated platform with multiple data modalities perfect defect materials is still lacking. This work...
We are developing a procedure for the quick identification of structural phases in thin film composition spread experiments which map large fractions compositional phase diagrams ternary metallic alloy systems. An in-house scanning x-ray microdiffractometer is used to obtain spectra from 273 different compositions on single library. A cluster analysis software then sort into groups order rapidly discover distribution diagram. The most representative pattern each group compared database known...
Abstract The Joint Automated Repository for Various Integrated Simulations (JARVIS) is an integrated infrastructure to accelerate materials discovery and design using density functional theory (DFT), classical force-fields (FF), machine learning (ML) techniques. JARVIS motivated by the Materials Genome Initiative (MGI) principles of developing open-access databases tools reduce cost development time discovery, optimization, deployment. major features are: JARVIS-DFT, JARVIS-FF, JARVIS-ML,...
Abstract Applied machine learning has rapidly spread throughout the physical sciences. In fact, learning-based data analysis and experimental decision-making have become commonplace. Here, we reflect on ongoing shift in conversation from proving that can be used, to how effectively implement it for advancing materials science. particular, advocate a big large-scale computations mentality model-oriented approach prioritizes use of support ecosystem computational models measurements. We also...
Abstract Science is and always has been based on data, but the terms ‘data-centric’ ‘4th paradigm’ of materials research indicate a radical change in how information retrieved, handled performed. It signifies transformative shift towards managing vast data collections, digital repositories, innovative analytics methods. The integration artificial intelligence its subset machine learning, become pivotal addressing all these challenges. This Roadmap Data-Centric Materials explores fundamental...
Here we present the results of using techno-economic analysis as constraints for machine learning guided studies new metal hydride materials.
On the basis of a set machine learning predictions glass formation in Ni–Ti–Al system, we have undertaken high-throughput experimental study that system. We utilized rapid synthesis followed by structural and electrochemical characterization. Using this dual-modality approach, are able to better classify amorphous portion library, which found be with full width at half maximum (fwhm) >0.42 Å–1 for first sharp X-ray diffraction peak. Proper phase labeling is important future efforts....
We have fabricated a series of composition spreads consisting ferroelectric BaTiO3 and piezomagnetic CoFe2O4 layers varying thicknesses modulated at nanometer level in order to explore artificial magnetoelectric thin-film heterostructures. Scanning microwave microscopy scanning superconducting quantum interference device were used map the dielectric magnetic properties as function continuously changing average across spreads, respectively. Compositions middle found exhibit ferromagnetism...