Crystal Structure Prediction via Deep Learning

Abstract

We demonstrate the application of deep neural networks as a machine-learning tool for the analysis of a large collection of crystallographic data contained in the crystal structure repositories. Using input data in the form of multiperspective atomic fingerprints, which describe coordination topology around unique crystallographic sites, we show that the neural-network model can be trained to effectively distinguish chemical elements based on the topology of their crystallographic environment. The model also identifies structurally similar atomic sites in the entire data set of ∼50000 crystal structures, essentially uncovering trends that reflect the periodic table of elements. The trained model was used to analyze templates derived from the known crystal structures in order to predict the likelihood of forming new compounds that could be generated by placing elements into these structural templates in a combinatorial fashion. Statistical analysis of predictive performance of the neural-network model, which was applied to a test set of structures never seen by the model during training, indicates its ability to predict known elemental compositions with a high likelihood of success. In ∼30% of cases, the known compositions were found among the top 10 most likely candidates proposed by the model. These results suggest that the approach developed in this work can be used to effectively guide the synthetic efforts in the discovery of new materials, especially in the case of systems composed of three or more chemical elements.

Perspectives on the Impact of Machine Learning, Deep Learning, and Artificial Intelligence on Materials, Processes, and Structures Engineering

“…These practices too are seeing significant benefit from current ML/AI tools. Some of these advancements were recently summarized [108][109][110][111][112][113][114][115][116][117]. However, the possibilities for materials compositions, microstructure, and architectures are vast-beyond human capacity alone to comprehensively search, discover, or design.…”

I2 Robot

“…In 2018, Kevin Ryan, Jeff Lengyel and Michael Shatruk at Florida State University in Tallahassee, US, trained a neural network based on the structural data of more than 50,000 known inorganic compounds. 2 They did not provide the system with any kind of chemical knowledge or understanding, such as the nature of the chemical bond or the arrangement of orbitals. The program was left to its own devices to look at the existing structures and derive its own view on how atoms combine to form materials.…”

Mechanistically Driven Control over Cubane Oxo Cluster Catalysts

“…In sharp contrast, their formation mechanisms and truly predictive syntheses remain grand challenges of inorganic chemistry. Considerable progress has been reported in computer-aided solid state and organic synthesis, [1][2][3] and major progress in synthetic coordination chemistry is expected from forthcoming machine learning approaches. Polynuclear transition metal-oxo complexes and polyoxometalates are prominent targets in current catalytic, bio-inorganic, medicinal and materials research.…”