AI News, Machine learning could be key to producing stronger, less corrosive metals

Machine learning could be key to producing stronger, less corrosive metals

But while researchers have studied grain boundaries for decades and gained some insight into the types of properties grain boundaries produce, no one has been able to nail down a universal system to predict if a certain configuration of atoms at grain boundaries will make a material stronger or more pliable.

student (Conrad Rosenbrock) and two professors -- one engineer (Eric Homer) and one physicist (Gus Hart) -- might have cracked the code by juicing a computer with an algorithm that allows it to learn the elusive 'why' behind the boundaries' qualities.

Their method, published in the most recent issue of Nature journal Computational Materials, provides a technique to produce a 'dictionary' of the atomic building blocks found in metals, alloys, semiconductors and other materials.

Their machine learning approach analyzes Big Data (think: massive data sets of grain boundaries) to provide insight into physical structures that are likely associated with specific mechanisms, processes and properties that would otherwise be difficult to identify.

The researchers believe they are at the front end of what could be a 10 or even 20-year process to create innovative alloy structures that provide practical solutions to major structures.

BYU researchers develop method that could produce stronger, more pliable metals

But while researchers have studied grain boundaries for decades and gained some insight into the types of properties grain boundaries produce, no one has been able to nail down a universal system to predict if a certain configuration of atoms at grain boundaries will make a material stronger or more pliable.

Their machine learning approach analyzes Big Data (think: massive data sets of grain boundaries) to provide insight into physical structures that are likely associated with specific mechanisms, processes and properties that would otherwise be difficult to identify.

“With Big Data models you lose some precision, but we’ve found it still provides strong enough information to connect the dots between a boundary and a property.” When it comes to metals, the process can evaluate properties like strength, weight and lifespan of materials, leading to the eventual optimization of the best materials.

We have a large database, and our algorithm is taking grain boundaries and comparing it against that database to connect them to certain properties.” The end goal is to make it easier and more efficient to develop materials that can be combined to make strong, lightweight and corrosion-free metals.

Discovering the building blocks of atomic systems using machine learning: application to grain boundaries

For small data sets, ASR does slightly better in predicting energy and temperature-dependent mobility trend;

In Figs.5 and 6, we plot some of the most important environments for determining whether a grain boundary will exhibit thermally activated mobility or not (Fig.5) or thermally damped mobility or not (Fig.6).

These most important LAEs are classified as such because their presence or absence in any of the GBs in the entire data set is highly correlated with the decision to classify them as thermally activated or not, or thermally damped or not.

GBs with partial dislocations emerging from the structure have been associated with thermally activated mobility and immobility, depending upon their presence in simple or complex GB structures;34 in addition, these structures have also been associated with shear coupled motion or the lack thereof.

However, the CSP cannot be directly compared with the LAE because CSP examines only nearest neighbors while the LAE encompasses a larger environment, including the defect at the edge of the LAE.47 Most importantly, this structure may not be immediately identified with any known metallic defect, but we know that it is highly correlated with thermally activated mobility across all the GBs in the data set.

The physical nature of those LAEs that we already understand suggests that these are the building blocks underlying important physical properties and that we may be on the precipice of understanding the atomic building blocks of GBs.

In addition to learning useful physical properties, the models provide access to a finite set of physical building blocks that are correlated with those properties throughout the high-dimensional GB space.

The work shows that analyzing big data regarding materials science problems can provide insight into physical structures that are likely associated with specific mechanisms, processes, and properties but which would otherwise be difficult to identify.

These methods may also provide a route to connect the crystallographic and atomic structure spaces so that existing expertize in the crystallographic space can be further optimized atomistically or vice versa.

While this is exciting within grain boundary science, the methodology presented here (and the SOAP descriptor in particular) has general applicability for building order parameters while studying changes that involve local structure.

Elizabeth A. Holm

Rohrer,“Grain boundary energies in body-centered cubic materials,”

Olmsted, “Phenomenology of shear-coupled grain boundary motion in symmetric tilt and general grain boundaries,”

Holm, “Incorporating atomistic models of lattice friction into BCC crystal plasticity models,”

Holm, “A hybrid simulation methodology for modeling dynamic recrystallization in UO2 LWR nuclear fuels,”

Holm, “Multiscale modeling of low temperature deformation in BCC metals,”

Rollett, “Recrystallized grain size in single phase materials,”

Olmsted, “Validating computed grain boundary energies in FCC metals using the grain boundary character distribution,”

Foiles, “How Grain Growth Stops: A mechanism for grain growth stagnation in pure materials,” Science 328 1138-1141 (2010) doi: 10.1126/science.1187833.

Holm, “Survey of grain boundary properties in FCC metals: I.

Foiles, “Survey of grain boundary properties in FCC metals: II.

Understanding why materials fail

Society’s reliance on the properties of key components in critical structures – made up of metal alloys and ceramic materials – is without question.

Crystallographies Dr Patala and his team investigated how crystals come together to form a material and how its structure influences properties like rate of diffusion, corrosion resistance, conductivity, inter-granular cracking, resistance to failure and the impact of extreme environments.

Cracks in these components can form at high temperature but their rate of formation remains unclear This was the challenge that Dr Patala set himself and his team: to create a reduced-order mathematical model that predicted how polycrystalline materials would perform and discover how grain boundaries would impact their ultimate strength, toughness and performance.

Dr Patala chose to use the mathematical properties of the well-established Voronoi network, a method of partitioning space, to automatically identify the network of three-dimensional polyhedra (whose vertices or corners represent the atoms) that are present in the structure of a grain boundary.

Dr Patala’s team also developed a pattern-matching technique that allows the comparison of the polyhedra found in the grain boundaries to a pre-existing database of hard-sphere packings (the way that atoms arrange themselves in a model material system).

Dr Patala and his team were then able to analyse the voids in the grain boundaries using the three-dimensional polyhedral geometries While Dr Patala’s research has focused on the analysis of grain boundaries in aluminium, his findings are applicable to most metals, ionic solids and some ceramics.

For example, identifying the voids in the grain boundary would help understand how small solute atoms interpose themselves within the grain boundary, influencing the ultimate strength and toughness of structural materials (e.g., through hydrogen embrittlement).

The developed mathematical model highlights a link between analysed grain boundary structures and those observed in metallic glasses, paving a way to evaluate grain boundaries using the theories proposed for amorphous materials.

Audio Media Preservation Through Imaging Conference (Day 1)

The Library of Congress hosted scientists and preservationists from around the world at a first-of-its-kind conference exploring new technologies that foster the ...