Artificial neural networks--algorithms inspired by connections in the brain--have "learned" to perform a variety of tasks, from pedestrian detection in self-driving cars, to analyzing medical images, to translating languages. Now, researchers at the University of California San Diego are training artificial neural networks to predict new stable materials.
"Predicting the stability of materials is a central problem in materials science, physics and chemistry," said senior author Shyue Ping Ong, a nanoengineering professor at the UC San Diego Jacobs School of Engineering. "On one hand, you have traditional chemical intuition such as Linus Pauling's five rules that describe stability for crystals in terms of the radii and packing of ions. On the other, you have expensive quantum mechanical computations to calculate the energy gained from forming a crystal that have to be done on supercomputers. What we have done is to use artificial neural networks to bridge these two worlds."
By training artificial neural networks to predict a crystal's formation energy using just two inputs--electronegativity and ionic radius of the constituent atoms--Ong and his team at the Materials Virtual Lab have developed models that can identify stable materials in two classes of crystals known as garnets and perovskites. These models are up to 10 times more accurate than previous machine learning models and are fast enough to efficiently screen thousands of materials in a matter of hours on a laptop. The team details the work in a paper published Sept. 18 in Nature Communications.
"Garnets and perovskites are used in LED lights, rechargeable lithium-ion batteries, and solar cells. These neural networks have the potential to greatly accelerate the discovery of new materials for these and other important applications," noted first author Weike Ye, a chemistry Ph.D. student in Ong's Materials Virtual Lab.
The team has made their models publicly accessible via a web application at http://crystals.ai. This allows other people to use these neural networks to compute the formation energy of any garnet or perovskite composition on the fly.
The researchers are planning to extend the application of neural networks to other crystal prototypes as well as other material properties.
Deep neural networks for accurate predictions of crystal stability
Weike Ye, Chi Chen, Zhenbin Wang, Iek-Heng Chu & Shyue Ping Ong
Nature Communications, vol.9, no.3800 (2018)
DOI: 10.1038%2Fs41467-018-06322-x
Contact information:
Shyue Ping Ong
UC San Diego Jacobs School of Engineering
s2ong@ucsd.edu
Tel: 858-534-2668
URL: https://materialsvirtuallab.org/
UC San Diego
#ArtificalIntelligence #MaterialScience #Nanocrystals
A eutectic mixture of methylammonium chloride and a zinc halide melts at about 100 degrees Celsius, forming a transient liquid layer at perovskite grain boundaries that lets crystals merge into larger, cleaner grains. The resulting films reached a carrier lifetime of roughly 295 microseconds and a surface recombination velocity of 0.3 cm/s, the lowest reported for lead halide perovskites.
Researchers at the University of Sydney have directly observed hydrogen dissolved in the interstitial spaces between metal atoms in alloys, a world first. Using cryogenic atom probe tomography with deuterium tracers, electron microscopy and theoretical calculations, the team showed hydrogen occupies sites between atoms rather than only collecting at defects, and that local atomic arrangement and chemistry dictate where it prefers to reside.
Researchers have developed carbon dots whose fluorescence shifts across the visible spectrum, from blue to red, depending on concentration, liquid environment and acidity. By tuning the particles' surfaces, the team produced versions that change color reversibly as pH shifts, then incorporated them into polymer powders, molded shapes and flexible films.
This week in nanotech
$13.6M raised
3 funding rounds · 20 deals tracked
New products
Ti-WCMBN Nanomaterialia Featured
DLS-1 Mass Spectrometer for Fusion Gas AnalysisHiden Analytical Featured
GHK-CuPeoples PepsMarket insights
Last 30 days: Quantum Technologies was the busiest vertical: 61 of 192 events (32%).
Money raised per month
