Abstract

Crystal growth is an important topic in materials science, chemical manufacturing, biology and even medicine. Much of our molecular scale understanding of crystal growth is anchored in classical nucleation theory. In this simple picture, solid clusters spontaneously shrink and grow until a single crystallite reaches a critical radius, and is subsequently able to reach macroscopic size. Extended theories for polymorphism, such as Ostwald’s rule of stages or the conjecture of Stranski and Totomanow have proven successful in some simple systems.

Despite these successes, an increasingly large body of evidence suggests that nature can exploit non-classical pathways to control crystallisation in ways that cannot be achieved synthetically. For example growth via amorphous precursors and evidence of “stable pre-nucleation clusters” is reported in a number of mineral systems.

This talk will discuss a number of ways in which computer simulation can improve our understanding of non-classical crystal growth. In particular, methods for overcoming the timescale limitations of molecular dynamics will be discussed, briefly illustrated with studies of ice nucleation, and various examples from biomineralisation – the mechanism by which nature manufactures shells, teeth and bone. Simulations will also be presented which demonstrate that nucleation barriers can emerge entirely as a consequence of pH, with no requirement for an interfacial free energy penalty between the growing crystal and its surroundings.