Deep learning algorithm to speed up materials discovery in emerging tech industries
Project to generate new hypothetical materials

Date:
August 25, 2020
Source:
University of South Carolina
Summary:
Solid-state inorganic materials are critical to the growth and
development of electric vehicle, cellphone, laptop battery and
solar energy technologies. However, finding the ideal materials with
the desired functions for these industries is extremely challenging.



FULL STORY ========================================================================== Solid-state inorganic materials are critical to the growth and
development of electric vehicle, cellphone, laptop battery and solar
energy technologies.

However, finding the ideal materials with the desired functions for
these industries is extremely challenging. Jianjun Hu, an associate
professor of computer science at the University of South Carolina is
the lead researcher on a project to generate new hypothetical materials.


==========================================================================
Due to the vast chemical design space and the high sparsity of candidates, experimental trials and first-principle computational simulations cannot
be used as screening tools to solve this problem. Instead, researchers
have developed a deep learning-based smart algorithm that uses a technique called generative adversarial network (GAN) model to dramatically improve
the material search efficiency up to two orders of magnitude. It has the potential to greatly speed up the discovery of novel functional materials.

The work, published in NPJ Computational Materials, was a collaboration
between researchers at the University of South Carolina College of
Engineering and Computing and Guizhou University, a research university
located in Guiyang, China.

Inspired by the deep learning technique used in Google's AlphaGo,
which learned implicit rules of the board game Go to defeat the game's
top players, the researchers used their GAN neural network to learn
the implicit chemical composition rules of atoms in different elements
to assemble chemically valid formulas. By training their deep learning
models using the tens of thousands of known inorganic materials deposited
in databases such as ICSD and OQMD, they created a generative machine
learning model capable of generating millions of new hypothetical
inorganic material formulas.

"There is almost an infinite number of new materials that could exist,
but they haven't been discovered yet," said Jianjun Hu. "Our algorithm,
it's like a generation engine. Using this model, we can generate a lot
of new hypothetical materials that have very high likelihoods to exist." Without explicitly modeling or enforcing chemical constraints such as
charge neutrality and electronegativity, the deep learning-based smart algorithm learned to observe such rules when generating millions of hypothetical materials' formulas. The predictive power of the algorithm
has been verified both by known materials and recent findings in materials discovery literature.

"One major advantage of our algorithm is the high validity, uniqueness
and novelty, which are the three major evaluation metrics of such
generative models," said Shaobo Li, a professor at Guizhou University
who was involved in this study.

This is not the first time that an algorithm has been created for
materials discovery. Past algorithms were also able to produce millions
of potential new materials. However, very few of the materials discovered
by these algorithms were synthesizable due to their high free energy and instability. In contrast, nearly 70 percent of the inorganic materials identified by Hu's team are very likely to be stable and then possibly synthesizable.

"You can get any number of formula combinations by putting elements'
symbols together. But it doesn't mean the physics can exist," said
Ming Hu, an associate professor of mechanical engineering at UofSC
also involved in the research. "So, our algorithm and the next step,
structure prediction algorithm, will dramatically increase the speed to screening new function materials by creating synthesizable compounds."
These new materials will help researchers in fields such as electric
vehicles, green energy, solar energy and cellphone development as they continually search for new materials with optimized functionalities. With
the current materials discovery process being so slow, these industries'
growth has been limited by the materials available to them.

The next major step for the team is to predict the crystal structure of
the generated formulas, which is currently a major challenge. However,
the team has already started working on this challenge along with several leading international teams. Once solved, the two steps can be combined
to discover many potential materials for energy conversion, storage and
other applications.

About University of South Carolina: The University of South Carolina
is a globally recognized, high-impact research university committed to
a superior student experience and dedicated to innovation in learning,
research and community engagement. Founded in 1801, the university offers
more than 350 degree programs and is the state's only top- tier Carnegie Foundation research institution. More than 50,000 students are enrolled at
one of 20 locations throughout the state, including the research campus
in Columbia. With 56 nationally ranked academic programs including top-
ranked programs in international business, the nation's best honors
college and distinguished programs in engineering, law, medicine, public
health and the arts, the university is helping to build healthier,
more educated communities in South Carolina and around the world.


========================================================================== Story Source: Materials provided by University_of_South_Carolina. Note:
Content may be edited for style and length.


========================================================================== Journal Reference:
1. Yabo Dan, Yong Zhao, Xiang Li, Shaobo Li, Ming Hu, Jianjun
Hu. Generative
adversarial networks (GAN) based efficient sampling of chemical
composition space for inverse design of inorganic materials. npj
Computational Materials, 2020; 6 (1) DOI: 10.1038/s41524-020-00352-0 ==========================================================================

Link to news story: https://www.sciencedaily.com/releases/2020/08/200825110638.htm

--- up 1 day, 6 hours, 50 minutes
* Origin: -=> Castle Rock BBS <=- Now Husky HPT Powered! (1337:3/111)