Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Oct 26;18(1):756-765.
doi: 10.1080/14686996.2017.1378060. eCollection 2017.

Machine learning reveals orbital interaction in materials

Tien Lam Pham  1   2 , Hiori Kino  2   3 , Kiyoyuki Terakura  1   3 , Takashi Miyake  2   3   4 , Koji Tsuda  3   5   6 , Ichigaku Takigawa  7   8 , Hieu Chi Dam  1   3   7
Affiliations

Machine learning reveals orbital interaction in materials

Tien Lam Pham et al. Sci Technol Adv Mater. .

Abstract

We propose a novel representation of materials named an 'orbital-field matrix (OFM)', which is based on the distribution of valence shell electrons. We demonstrate that this new representation can be highly useful in mining material data. Experimental investigation shows that the formation energies of crystalline materials, atomization energies of molecular materials, and local magnetic moments of the constituent atoms in bimetal alloys of lanthanide metal and transition-metal can be predicted with high accuracy using the OFM. Knowledge regarding the role of the coordination numbers of the transition-metal and lanthanide elements in determining the local magnetic moments of the transition-metal sites can be acquired directly from decision tree regression analyses using the OFM.

Keywords: Material descriptor; data mining; machine learning; magnetic materials; material informatics.

PubMed Disclaimer

Conflict of interest statement

No potential conflict of interest was reported by the authors.

Figures

None
Graphical abstract
Figure 1.
Figure 1.
OFM representation for an Na atom in a regular octahedral site surrounded by six Cl atoms: atomic one-hot vector for Na (left), representation for the six Cl atoms surrounding the Na atom (middle), and representation for the Na atom surrounded by six Cl atoms (right).
Figure 2.
Figure 2.
Decision tree regression for Mn (a), Fe (b), Co (c), and Ni (d). In each leaf, the upper part indicates the values of the local magnetic moments, whereas the lower part indicates the number of positive (P) and negative (N) examples.
Figure 3.
Figure 3.
Comparison of formation energies calculated using DFT and those predicted through machine learning (ML-predicted), using OFM.
Figure 4.
Figure 4.
Standard deviations of local OFMs of QM7 (a) and LATX (b) datasets.
See this image and copyright information in PMC

References

    1. Yousef S, Da G, Thanh N, et al. . Data mining for materials: computational experiments with compounds. Phys Rev B. 2012;85:104104.
    1. Yang S, Lach-hab M, Vaisman II, et al. . Identifying zeolite frameworks with a machine learning approach. Phys Chem C. 2009;113:21721–21725.
    1. Hautier G, Fischer CC, Jain A, et al. . Finding nature’s missing ternary oxide compounds using machine learning and density functional theory. Chem Mater. 2010;22:3762–3767.
    1. Snyder JC, Rupp M, Hansen K, et al. . Finding density functionals with machine learning. Phys Rev Lett. 2012;108:253002. - PubMed
    1. Isayev O, Fourches D, Muratov EN, et al. . Materials cartography: representing and mining materials space using structural and electronic fingerprints. Chem Mater. 2015;27:735–743.