IISc SSCU Webmail (Bagchi group)

Nucleation and growth

Understanding the mechanism of nucleation of the stable phase inside the metastable parent phase during a first order phase transition has been a subject of outstanding interest in natural science. The problem becomes even more challenging as the spinodal is approached and here the mechanism of nucleation is not clearly understood, with many questions regarding the existence of a free energy barrier and the non-zero value of the surface tension have remained unanswered.

Very recently, we have undertaken extensive computer simulation studies to probe the molecular mechanism for the onset of instability. We have constructed the multidimensional free energy surface of nucleation as a function of multiple reaction coordinates, both for supercooled Lennard-Jones fluid and for 2- and 3-dimensional Ising models. While the classical Becker-Döring (BD) picture of homogenous nucleation, that assumes the growth of a single nucleus by single particle addition, holds good at low to moderate supersaturation, the formation of the new stable phase becomes more collective and spread over the whole system at large supersaturation. As the spinodal curve is approached from the coexistence line, the free energy, as a function of the size of the largest liquid-like cluster, develops a minimum at a sub-critical cluster size. This minimum at intermediate size is responsible for the barrier towards further growth of the nucleus at large supersaturation. As the spinodal is approached closely, this minimum gradually disappears and so does the free energy barrier for the cluster growth.  We find the emergence of an alternative free energy pathway (with a barrier less than that in the BD picture) that involves participation of many sub-critical liquid-like clusters and the growth is promoted by coalescence with intermediate sized clusters present in the neighbourhood of the largest cluster.  Very close to the spinodal the free energy surface becomes quite flat, the significance of a critical nucleus is lost and the classical Becker-Döring picture of nucleation breaks down. 

Reference: http://arxiv.org/abs/cond-mat/0702158