Revolution in a beaker
Researchers at the University of Amsterdam, the University of Vienna and the Foundation for Fundamental Research on Matter (FOM) have produced unexpected results using advanced computer simulations based on classical nucleation theory. This elegant theory predicts that seed clusters can only grow if they have a critical size, but the predictions are not always correct. FOM workgroup leader Professor dr. Peter Bolhuis and his colleagues have now found the cause: the nucleation clusters are not homogenous, but are structured in layers as in an onion. The innermost cluster has the most stable crystal structure but is surrounded by an indecisive cloud of particles. The combination of the size of the cluster and the size of the surface cloud is an accurate predictor of whether nucleation will take place. The researchers published their results online on 22 February 2011 in Physical Review Letters and the article has been highlighted as suggested reading by the journal's editors.
Nucleation theory gives accurate predictions and due to its generic nature, it can be applied to nucleation processes during phase transitions in a wide variety of different systems. Small seed clusters will always be dissolved, and these will only grow further if they happen to become big enough. Previous simulations had revealed that small seed clusters that had been assumed to shrink did, in fact, grow, whereas other seed clusters that appeared to be large, disappeared. This meant that there had to be other properties of the seed clusters which determine their fate.
The researchers have now demonstrated that the nucleation clusters are not homogenous but are structured in layers. Only the innermost cluster has the desired most stable crystalline structure. This cluster is surrounded by an indecisive cloud of particles which, although ordered, are not really crystalline. If both the size of the cluster and the size of the surface cloud are included in the simulation model then it can be accurately predicted whether nucleation and the associated phase transition will take place. If the cloud of particles is big enough then even small seed clusters that cannot grow by themselves can lead to the desired phase transition.
Whether or not the new phase is formed is therefore down to the indecisive cloud of particles on the surface. These insights have provided us with a better understanding of a crucial phase of the revolution in the beaker: the initial phase.
See it for yourself
Have a go at the following experiment: take a beaker or a plastic bottle of water (it works even better with beer in a transparent glass bottle) and place this upright in the freezer for the night. Most people are amazed to find that the water is still liquid the next morning and has not frozen. Only when you give the bottle a brief shock does the water rapidly freeze. What exactly has happened to the water? First of all, the water was cooled to its freezing point. It did not freeze immediately and remained liquid even after it had cooled even further still. This so-called undercooled liquid state is the starting point for the revolution of the water molecules. They are dissatisfied so to speak and want to form ice instead of water. This situation can persist for a very long time. Only after a brief shock can a small number of the water molecules spontaneously form a tiny cluster of the ice crystal, which then rapidly grows and eventually takes over the entire water bottle. Pressure waves that develop in the bottle due to the shock temporarily reduce the energy barrier for nucleation. Consequently, the cloud of indecisive particles immediately decides to freeze.
Reference
The article.
DOI: 10.1103/PhysRevLett.106.085701
Contact
Prof.dr. Peter Bolhuis (020) 525 64 47.