Berlin 2024 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

DY: Fachverband Dynamik und Statistische Physik

DY 12: Statistical Physics far from Thermal Equilibrium II

DY 12.5: Talk

Monday, March 18, 2024, 16:00–16:15, BH-N 334

Hard sphere nucleation revisited: bringing simulation and experiment together — Sahana Kale and •Hans Joachim Schöpe — Universität Tübingen, Institut für angewandte Physik, Auf der Morgenstelle 10, 72076 Tübingen

Crystallization of a metastable melt is one of the most important non-equilibrium phenomena in classical physics. A milestone in crystal nucleation studies was the first prediction of absolute nucleation rate densities (NRDs) in hard sphere (HS) systems using the combination of computer simulations and classical nucleation theory (CNT), published in Nature in 2001. This work has been complemented in recent years by more advanced simulation work. However, a direct comparison of the experimental NRDs with those from the simulations shows a highly unsatisfactory result: the shape of the curves are qualitatively different and experimental and theoretical data diverge by up to 18 orders of magnitude!

Using laser-scanning confocal microscopy we study crystal nucleation in colloidal HS. We follow and characterise the formation of individual crystallites. Furthermore, we determine NRDs and critical radii in two different ways: 1) directly from the data without recourse to any model and 2) within the framework of CNT. The direct data analysis reproduces the existing experimental data sets from Bragg and small angle scattering, while the CNT-based analysis reproduces the simulation data. However, as mentioned above, the two data sets are incompatible. Accordingly, the cause of the huge discrepancy is due to shortcomings of the CNT and simulation techniques.

Keywords: crystal nucleation; hard spheres; experiment; confocal microscopy; classical nucleation theory