SKM 2023 – wissenschaftliches Programm

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

CPP: Fachverband Chemische Physik und Polymerphysik

CPP 19: Crystallization, Nucleation and Self-Assembly

CPP 19.4: Vortrag

Dienstag, 28. März 2023, 10:30–10:45, MER 02

Crystallization precursors in polymer melt analyzed by machine learning — •Atmika Bhardwaj^{1, 2}, Marco Werner¹, and Jens-Uwe Sommer^{1, 2} — ¹Leibniz-Institut für Polymerforschung Dresden e. V., Hohe Str. 6, D-01069 Dresden, Germany — ²Institute for Theoretical Physics, Technische Universität Dresden, Zellescher Weg 17, D-01069 Dresden, Germany

Crystallization in polymers is a long-standing problem in both experimental and theoretical polymer science. The transition dynamics occurring in an under-cooled polymer melt is a local environmental phenomenon rather than a property of individual particles (or monomers) and depends on subtle conformation patterns such as entanglements between the chains. We develop machine learning (ML) methods to study this non-equilibrium thermodynamic process. Upon recognizing the relevant parameter set to explore different phases during polymer crystallization, we investigate the spatial and temporal patterns of the precursor states that determine the nucleation sites. The objective is to recognize the precursors that stimulate crystal growth before the occurrence of such development.