The description of nucleation has been dominated by the classical nucleation theory (CNT) for more than 80 years. However, significant problems exist for the understanding and description of real nucleation or crystallization events using CNT. Most problematic in CNT is certainly the assumption of bulk properties even for the smallest metastable clusters. Instead, it has turned out for CaCO3 and several other systems that nucleation can also follow an alternative pathway involving prenucleation clusters, which are thermodynamically stable with respect to the initial ions and are solutes containing some tens of ions. It turned out that for CaCO3, the prenucleation clusters can be understood as dynamic polymers and their formation is entropically driven by release of ion hydration water. Prenucleation clusters even form in undersaturated solution without driving force towards precipitation. The current knowledge about prenucleation clusters will be presented and discussed.
But even beyond the stage of nucleation, crystal growth does not necessarily proceed along the classical mechanism of the addition of atom/ion/molecule building units to the growth sites of a growing crystal. Instead, nanocrystals can self assemble via different mechanisms or crystallize from an amorphous precursor phase to form so called "Mesocrystals". This is an abbreviation for "Mesoscopically structured crystals" and involves the crystallographically aligned superstructure of nanoparticles as a kinetic intermediate. Potential fusion of the already crystallographically aligned nanoparticles after displacement of surface attached stabilizer molecules can lead to crystallographic fusion of the two aligned crystal faces better known as "Oriented Attachment". These processes are known as "Nonclassical Crystallization". Several of these nonclassical mechanisms will be presented and discussed.
helmut.coelfen@uni-konstanz.de
