SMT-Aided Combinatorial Materials Discovery

Ermon, Stefano; Bras, Ronan Le; Gomes, Carla P.; Selman, Bart; Dover, R. B. van

doi:10.1007/978-3-642-31612-8_14

Cited by 21 publications

(22 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There have been a number of efforts directed at developing algorithms that allow rapid identification of key trends and features in diffraction data taken across multiphase CCS libraries, and subsequent attribution of such data to individual phases. [400][401][402][403][404][405][406][407] By performing cluster analysis, similar diffraction patterns, even among hundreds of spectra, can be rapidly identified. Similar XRD patterns usually indicate the presence of similar phases in a particular composition region.…”

Section: High-throughput Mapping Of Phase Diagramsmentioning

confidence: 99%

Applications of high throughput (combinatorial) methodologies to electronic, magnetic, optical, and energy-related materials

Green

Takeuchi

Hattrick‐Simpers

2013

Journal of Applied Physics

227

189

View full text Add to dashboard Cite

High throughput (combinatorial) materials science methodology is a relatively new research paradigm that offers the promise of rapid and efficient materials screening, optimization, and discovery. The paradigm started in the pharmaceutical industry but was rapidly adopted to accelerate materials research in a wide variety of areas. High throughput experiments are characterized by synthesis of a "library" sample that contains the materials variation of interest (typically composition), and rapid and localized measurement schemes that result in massive data sets. Because the data are collected at the same time on the same "library" sample, they can be highly uniform with respect to fixed processing parameters. This article critically reviews the literature pertaining to applications of combinatorial materials science for electronic, magnetic, optical, and energy-related materials. It is expected that high throughput methodologies will facilitate commercialization of novel materials for these critically important applications. Despite the overwhelming evidence presented in this paper that high throughput studies can effectively inform commercial practice, in our perception, it remains an underutilized research and development tool. Part of this perception may be due to the inaccessibility of proprietary industrial research and development practices, but clearly the initial cost and availability of high throughput laboratory equipment plays a role. Combinatorial materials science has traditionally been focused on materials discovery, screening, and optimization to combat the extremely high cost and long development times for new materials and their introduction into commerce. Going forward, combinatorial materials science will also be driven by other needs such as materials substitution and experimental verification of materials properties predicted by modeling and simulation, which have recently received much attention with the advent of the Materials Genome Initiative. Thus, the challenge for combinatorial methodology will be the effective coupling of synthesis, characterization and theory, and the ability to rapidly manage large amounts of data in a variety of formats. V C 2013 AIP Publishing LLC. [http://dx

show abstract

Section: High-throughput Mapping Of Phase Diagramsmentioning

confidence: 99%

Applications of high throughput (combinatorial) methodologies to electronic, magnetic, optical, and energy-related materials

Green

Takeuchi

Hattrick‐Simpers

2013

Journal of Applied Physics

227

189

View full text Add to dashboard Cite

show abstract

“…Domain knowledge is often given as a set of additional constraints [149]. The integration of additional domain knowledge generates a new hybrid formulation of the ML problem which then ideally leads to physically meaningful and significantly more accurate interpretations of the data [38]. Besides adding constraints, expert knowledge can be incorporated in different ways.…”

Section: State Of the Artmentioning

confidence: 99%

Machine learning for material characterization with an application for predicting mechanical properties

Stoll

Benner

2021

GAMM-Mitteilungen

View full text Add to dashboard Cite

Currently, the growth of material data from experiments and simulations is expanding beyond processable amounts. This makes the development of new data‐driven methods for the discovery of patterns among multiple lengthscales and time‐scales and structure‐property relationships essential. These data‐driven approaches show enormous promise within materials science. The following review covers machine learning (ML) applications for metallic material characterization. Many parameters associated with the processing and the structure of materials affect the properties and the performance of manufactured components. Thus, this study is an attempt to investigate the usefulness of ML methods for material property prediction. Material characteristics such as strength, toughness, hardness, brittleness, or ductility are relevant to categorize a material or component according to their quality. In industry, material tests like tensile tests, compression tests, or creep tests are often time consuming and expensive to perform. Therefore, the application of ML approaches is considered helpful for an easier generation of material property information. This study also gives an application of ML methods on small punch test (SPT) data for the determination of the property ultimate tensile strength for various materials. A strong correlation between SPT data and tensile test data was found which ultimately allows to replace more costly tests by simple and fast tests in combination with ML.

show abstract

“…Proposed clustering methods such as hierarchical clustering (HCA) (Long et al 2007), dynamic time warping kernel clustering (Le Bras et al 2011) and mean shift theory (Kusne et al 2014) produce maps of phase regions, but fail to resolve mixtures or identify basis patterns, and do not necessarily produce results consistent with physics. Constraint reasoning approaches include satisfiability modulo theory (SMT) methods (Ermon et al 2012), which can provide physically meaningful results, but depend heavily on effective pre-processing such as peak identification, and are computationally intensive. Approaches based on non-negative matrix factorization (NMF) (Long et al 2009) are computationally efficient, but generally perform poorly when peakshifting phenomena are present.…”

Section: High-throughput Characterizationmentioning

confidence: 99%

Phase-Mapper: An AI Platform to Accelerate High Throughput Materials Discovery

Xue

Bai

Bras

et al. 2016

Preprint

Self Cite

View full text Add to dashboard Cite

High-Throughput materials discovery involves the rapid synthesis, measurement, and characterization of many different but structurally-related materials. A key problem in materials discovery, the phase map identification problem, involves the determination of the crystal phase diagram from the materials composition and structural characterization data. We present Phase-Mapper, a novel AI platform to solve the phase map identification problem that allows humans to interact with both the data and products of AI algorithms, including the incorporation of human feedback to constrain or initialize solutions. Phase-Mapper affords incorporation of any spectral demixing algorithm, including our novel solver, Ag-ileFD, which is based on a convolutive non-negative matrix factorization algorithm. AgileFD can incorporate constraints to capture the physics of the materials as well as human feedback. We compare three solver variants with previously proposed methods in a large-scale experiment involving 20 synthetic systems, demonstrating the efficacy of imposing physical constrains using AgileFD. Phase-Mapper has also been used by materials scientists to solve a wide variety of phase diagrams, including the previously unsolved Nb-Mn-V oxide system, which is provided here as an illustrative example.

show abstract

SMT-Aided Combinatorial Materials Discovery

Cited by 21 publications

References 13 publications

Applications of high throughput (combinatorial) methodologies to electronic, magnetic, optical, and energy-related materials

Applications of high throughput (combinatorial) methodologies to electronic, magnetic, optical, and energy-related materials

Machine learning for material characterization with an application for predicting mechanical properties

Phase-Mapper: An AI Platform to Accelerate High Throughput Materials Discovery

Contact Info

Product

Resources

About