frenkel@amolf.nl
Crystal nucleation is a very common phenomenon, but it is not well understood. For almost a century, the theoretical models that predict the rate of crystal nucleation could not be tested in detail. That situation is now changing because of two developments in colloidal science: 1. better real-space probes of (colloidal) crystal nucleation and 2. better numerical modeling techniques. In my talk, I will review recent advances in our understanding of homogeneous and heterogeneous crystal nucleation and I will indicate some of the areas where our understanding is still far Confocal laser scanning microscopy was used to observe crystal nucleation in a colloidal suspension of nearly hard sphere particles. Direct imaging in three dimensions allowed identification and observation of nucleation and growth of crystalline regions. The structure and shape of subcritical and critical crystal nuclei were analyzed. Surprisingly, the surfaces of the nuclei were very rough, making them appear fractal in nature. The good time resolution that was accessible with the fast confocal scanner allowed us to follow the evolution of a large number of nuclei, to determine the critical nucleus size, and to estimate the nucleation density rate
. Upper and lower bounds for the solid/fluid interface surface energy are calculated from the estimate for , and these values are compared with the properties of very small subcritical
from complete.
Examples will include: colloidal crystals, molten salts and extra-terrestrial diamonds.