Tokyo – If the properties of materials can be reliably predicted, the process of developing new products for a wide variety of industries can be streamlined and accelerated. In a study published in XXX Advanced intelligent systems, researchers at the Tokyo University Institute of Industrial Sciences used nuclear loss spectroscopy to determine the properties of organic molecules using machine learning.
The spectroscopy techniques Energy Loss Near Edge Structure (ELNES) and X-ray Near Edge Structure (XANES) are used to determine information about the electrons and thus the atoms in materials. They have high sensitivity and high resolution and have been used to examine a range of materials from electronic devices to drug delivery systems.
However, the association of spectral data with the properties of a material – things like optical properties, electron conductivity, density, and stability – remains ambiguous. Machine learning (ML) approaches have been used to extract information for large complex data sets. Such approaches use artificial neural networks based on how our brains work to constantly learn to solve problems. Although the group previously used ELNES / XANES spectra and ML to obtain information about materials, the result was not related to the properties of the material itself. Therefore, the information could not be easily translated into developments.
Now the team has used ML to uncover information hidden in the simulated ELNES / XANES spectra of 22,155 organic molecules. âThe ELNES / XANES spectra of the molecules or, in this scenario, theirâ descriptors âwere then entered into the system,â explains first author Kakeru Kikumasa. âThis descriptor is something that can be measured directly in experiments and therefore can be determined with high sensitivity and resolution. This method is of great use for material development because it has the potential to reveal where, when and how certain material properties arise. “
A model created solely from the spectra was able to successfully predict the so-called intensive properties. However, he was unable to predict extensive properties that depend on the size of the molecules. In order to improve the prediction, the new model was constructed by taking into account the ratios of three elements in relation to carbon (which is present in all organic molecules) as additional parameters in order to correctly predict extensive properties such as molecular weight.
âOur ML learning treatment of core loss spectra enables an accurate prediction of extensive material properties such as internal energy and molecular weight. The connection between core loss spectra and extensive properties has never been established before; However, artificial intelligence was able to uncover the hidden connections. Our approach could also be used to predict the properties of new materials and functions, âsays senior author Teruyasu Mizoguchi. “We believe our model will be a very useful tool for high-throughput development of materials in a wide variety of industries.”
The study “Quantification of the Properties of Organic Molecules Using Core-Loss Spectra as Neural Network Descriptors” was published in Advanced intelligent systems at DOI: 10.1002 / aisy.202100103.
Via the Institute of Industrial Science (IIS), the University of Tokyo
Institute of Industrial Science (IIS) at the University of Tokyo is one of the largest university-linked research institutes in Japan.
The IIS comprises more than 120 research laboratories, each headed by a faculty member, with more than 1,200 members, including approximately 400 staff and 800 students who are actively involved in education and research. Our activities cover almost all areas of the engineering discipline. Since its inception in 1949, IIS has worked to fill the large gaps between academic disciplines and real world applications.
Advanced intelligent systems
Quantification of the properties of organic molecules using nuclear loss spectra as descriptors for neural networks
Publication date of the article
Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of press releases sent to EurekAlert! by contributing institutions or for the use of information via the EurekAlert system.