Understanding the nature of the structural evolution during nucleation and growth of crystal is still a fundamental and challenging issue in condensed-matter science. One of the major relevant topics is the Ostwald’s step rule, first proposed in 1897 as an empirical rule, stating that in general it is not the most stable, but the least stable, polymorph that crystallizes first.
However, Ostwald’s step rule is not yet a universal law due to two factors: Firstly, there are exceptions (although very few) to the rule. Secondly, an undisputed theoretical basis for the rule has not been formulated successfully. In over the past 100 years, various attempts were made to solve related problems, but none of them were successful. In 1978, on the basis of the mean-field treatment, Alexander and McTague published their striking result predicting that a body-centered cubic (bcc) structure should be formed first regardless of whether a more thermodynamically stable one exists, under certain conditions.
Researchers of the complex fluid group of our institute have made substantial progress in this subject. Taking colloidal crystal as a model system, they adopted a brand-new experimental method to improve the speed of diagnosing the crystal structure by three orders of magnitude and further found, for the first time, the existence of metastable BCC in the process of colloidal crystallization. Subsequently, they proposed a theoretical model to formulate the kinetics of the concurrent liquid-metastable and metastable-stable transitions. This model is well-supported by their observations. They found that when the ratio of the metastable-stable transition rate to the liquid-metastable rate is very large, the metastable state can become undetectable. This situation is rather analogous to a rabbit eating grass: grass grows from soil, and a rabbit grows by eating grass. If the eating rate is much higher than the grass growth rate, the grass will be too little to be seen, but it does not mean that grass never grows. Therefore, their findings provide a possible explanation for the exceptions to Ostwald’s step rule. They further kinetically demonstrated that the reason for the presence of the metastable phase is that its presence can provide an alternative, lower energy nucleation-growth pathway.
In addition, this group measured shear moduli variation in the metastable (BCC)-stable (FCC) structure transition of charged colloidal crystals by the combination techniques of torsional resonance spectroscopy and reflection spectrometer. They found that in the transition process, the moduli are much smaller than theoretical ones and this can be chalked up to crystalline imperfection in the transition state.
The relevant publications include (1) Langmuir (2015, DOI: 10.1021/acs.langmuir.5b00917); (2) J. Chem. Phys. (2015, DOI: 10.1063/1.4932684); (3) Phys. Rev. E 82, 010401 (2010, DOI: 10.1103/PhysRevE.82.010401) (Rapid Communications); (4) Langmuir 27, 7439 (2011, DOI: 10.1021/la200407h)；(5) Colloid Surfaces A, (2011, DOI: 10.1016/j.colsurfa.2010.11.051).
Figure: The structure composition-time curves of BCC and FCC with different volume fraction Φ. At low Φ: BCC is stable structure and the relevant phase transition process is liquid-BCC; With the increase of Φ, BCC becomes metastable structure and FCC becomes the stable one; Within a certain range of Φ, it can be seen that BCC and FCC phase coexistence and transition of Liquid-BCC-FCC; When Φ is further increased, the transition process is still Liquid-BCC-FCC, but the metastable state can become undetectable due to its extremely short life time.