Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

MM: Fachverband Metall- und Materialphysik

MM 18: SYMD contributed

MM 18.6: Vortrag

Mittwoch, 19. März 2025, 11:45–12:00, H23

Inferring Structure-Property Relationships with Artificial Intelligence: A Lignin Case Study — •Matthias Stosiek and Patrick Rinke — Technical University Munich, Munich, Germany

The potential of lignin as an abundant, underutilized biopolymer is increasingly being realized. A key challenge for the targeted production of lignins remains the poorly understood relation between lignin properties and its complex structure. Artificial intelligence (AI) methods could reveal such structure-function relationships but remain elusive in biomaterials research.

Structurally diverse lignins are extracted from birch wood combining the Aqua Solv Omni (AqSO) biorefinery process and AI-guided data acquisition[1]. Each lignin sample is characterized with 2D nuclear magnetic resonance (NMR) spectroscopy. A total of 95 collected NMR spectra are complemented with measurements of key lignin properties such as the antioxidant activity.

To establish structure-function relationships, we first correlate regions of the NMR spectra with the corresponding property measurements. Subsequently, we use RFR feature importance analysis to identify structural features that correlate with each property and provide a chemical interpretation of our findings. For instance, we find that a higher number of β-O-4 bonds leads to a lower surface tension in water indicating a more linear lignin structure. Our structure-inference approach is designed to be general and applicable to a wide range of materials and characterization data.

[1] D. Diment et al., ChemSusChem 2024, e202401711.

Keywords: Structure-Property Relation; Lignin Carbohydrate Complexes; Random Forest Regression; Nuclear Magnetic Resonance Spectroscopy; Machine Learning