ľ¹Ï¸£ÀûÓ°ÊÓ

Speaker: dr. Eduardo Sanz (Physical Chemistry Department of the ľ¹Ï¸£ÀûÓ°ÊÓ Complutense of Madrid, Spain)

Chair and host: Prof. dr. ir. Marjolein

Title: Addressing fundamental question in nucleation phenomena with Seeding simulations

Abstract: Understanding first order phase transitions is of great importance in many fields, ranging from biology, to atmospheric science, physics, geology or industry. In the absence of impurities or external surfaces, first order phase transitions start with the emergence of a nucleus of the stable phase in the bulk of the parent metastable phase. A nucleus is ``critical''  if it is big enough so that it has 50 per cent chances to either grow or redissolve. Although the emerging phase is more stable, the presence of an unfavourable interface between the nucleus and the parent phase can delay to a great extent the phase transition. Therefore, the surface tension, $\gamma$, or the free energy per unit area between both phases, plays a key role in the development of first order phase transitions. Whereas $\gamma$ can be readily measured for a flat interface at equilibrium --at least between fluid phases--it cannot be directly probed for curved interfaces, which is the relevant case for nucleation. Moreover, the fact that critical nuclei are nanoscopic objects makes it very difficult to observe them in experiments, let alone measuring their $\gamma$. Computer simulations do have access to the time and length scales relevant for the observation of critical nuclei.  However, whereas the methodology and theoretical framework for computing $\gamma$ for flat interfaces is very well established, that for curved interfaces is still under development. We have extensively developed in the past years the so-called Seeding method to study nucleation phenomena. This method consists in obtaining with simulations the properties of critical clusters and ``plug'' them in the Classical Nucleation Theory (CNT) formalism to obtain predictions of the nucleation rate and of the $\gamma$ curvature dependence. The Seeding approach is rather powerful, but it has to be carefully assessed with independent methods because it relies on an arbitrary definition of the interface location.

In this presentation I'll show how Seeding has helped to clarify fundamental questions in the nucleation of ice, colloidal crystals, drops and bubbles.