Artificial intelligence ARTIST instantly captures materials’ properties

New approach holds potential to slash research and development costs for designer materials and technologies of the future
Artificial intelligence ARTIST - Jari Järvi/Aalto University
Illustration: Jari Järvi/Aalto University

Researchers at Aalto University and the Technical University of Denmark have developed an artificial intelligence (AI) to seriously accelerate the development of new technologies from wearable electronics to flexible solar panels. ARTIST, which stands for Artificial Intelligence for Spectroscopy, instantly determines how a molecule will react to light—clinch-pin knowledge for creating the designer materials needed for tomorrow’s technology.

Scientists traditionally study molecular reactions to external stimuli with spectroscopy, a widely used method across the natural sciences and industry. Spectroscopy probes the internal properties of materials by observing their response to, for example, light, and has led to the development of countless everyday technologies. Existing experimental and computational spectroscopy approaches can be, however, incredibly costly. Time in highly specialised laboratories is expensive and often severely limited, while computations can be tedious and time-intensive.

With ARTIST, the research team offers a paradigm shift to how we determine the spectra—or response to light—of individual molecules.

‘Normally, to find the best molecules for devices, we have to combine previous knowledge with some degree of chemical intuition. Checking their individual spectra is then a trial-and-error process that can stretch weeks or months, depending on the number of molecules that might fit the job. Our AI gives you these properties instantly,’ says Milica Todorović, a postdoctoral researcher at Aalto University.

With its speed and accuracy, ARTIST has the potential to speed up the development of flexible electronics, including light-emitting diodes (LEDs) or paper with screen-like abilities. Complementing basic research and characterization in the lab, ARTIST may also hold the key to producing better batteries and catalysts, as well as creating new compounds with carefully selected colours.

The multidisciplinary team trained the AI in just a few weeks with a dataset of more than 132,000 organic molecules. ARTIST can predict with exceedingly good accuracy just how those molecules—and those similar in nature—will react to a stream of light. The team now hopes to expand its abilities by training ARTIST with even more data to make an even more powerful tool.

‘Enormous amounts of spectroscopy information sit in labs around the world. We want to keep training ARTIST with further large datasets so that it can one day learn continuously as more and more data comes in,’ explains Aalto University Professor Patrick Rinke.

The researchers aim to release ARTIST on an open science platform in 2019, and it is currently available for use and further training upon request.

The study was published in Advanced Science on 29 January 2019. 

Link to the open access article

Kunal Ghosh, Annika Stuke,  Milica Todorović, Peter Bjørn Jørgensen, Mikkel N. Schmidt, Aki Vehtari, Patrick Rinke (2019): Deep Learning Spectroscopy: Neural Networks for Molecular Excitation Spectra, Advanced Science. https://doi.org/10.1002/advs.201801367

More information

Professor Patrick Rinke
Aalto University
Telephone: +358504433199
Email: [email protected]

Postdoctoral researcher Milica Todorović 
Aalto University
Telephone: +358503310029
Email: [email protected]

 

One atom at a time

AI is not only changing the way we process data, it’s changing how we carry out research, period.

More on Rinke's and Todorović's work
Milica Todorovic and Patrick Rinke. Photographer: Venla Helenius.

Related news

Nanocellulose bicycle Photo: Eeva Suorlahti
Cooperation, Press releases, Research & Art Published:

Tomorrow's Sustainable Lifestyles Showcased in Otaniemi

One of Helsinki Design Week's main events, Designs for a Cooler Planet, will showcase Aalto University's cross-cutting future prototypes, such as a nanocellulose bicycle, microbial headphones and Ioncell clothes
Photoactive rod-like virus bundle schematic
Press releases, Research & Art, University Published:

Dyes and viruses create new composite material for photooxidation reactions

A recent study shows that native viruses can be employed as a scaffold to immobilise photoactive molecules to potentially oxidise organic pollutants present in wastewater, under visible light irradiation
Kuvituskuva, jonka pohjana käytetty visualisointia eduskuntavaaleihin liittyvistä aihetunnisteista
Press releases Published:

Finns were reluctant to share election tweets of bots

According to the final report of ELEBOT project, the effect of Twitter bots was minimal during this year’s Finnish elections
breakben aivomittausanturi kuvaaja marko havu
Press releases, Research & Art Published:

Breaking through barriers for a revolution in brain scans

The EU-funded BREAKBEN project promises better and more detailed brain scans
  • Published:
  • Updated:
Share
URL copied!