LLNL scientist Tuan Anh Pham. Credit: Blaise Douros / LLNL
Researchers from the Lawrence Livermore National Laboratory (LLNL) have developed a novel method to rapidly predict the chemical composition, and structure, of a material. The research lays the groundwork for the development of an automated method to characterize heterogeneous materials.
In the research, published in Chemistry of Materials, the team developed a method combining machine learning and X-ray absorption spectroscopy to allow for the speciation of amorphous carbon nitrides. The integration of machine learning and atomistic simulations allows researchers to establish correlations between the local atomic structures observed and their spectroscopic signatures. This correlation serves as the basis for interpreting experimental data generated by X-ray absorption spectroscopy analysis.
The method addresses long-standing challenges faced when analyzing materials such as the carbonaceous residues produced when explosives detonate. Traditional methods employed by materials scientists are often time-consuming and rely on the use of empirical parameters to achieve accurate analysis.
"Our approach not only enhances our understanding of these materials but also lays the groundwork for similar studies across different material systems and characterization methods. For example, the approach can be readily employed to predict elemental speciation for a broad range of carbonaceous residues and provide inputs for improving detonation models," said Tuan Anh Pham, principal investigator for the project.
The methods developed by the team offer a comprehensive way for materials scientists to analyze the atomic speciation of disordered systems. Additionally, the method provides the necessary versatility to be adapted to investigate additional classes of materials which could lead to future real-time analysis of spectroscopic data.