Prof Zhongyn Fan - Brunel University
Recent advances in the understanding of heterogeneous nucleation
BCAST, Brunel University, UK
Solidification starts with nucleation followed by growth. 60 years of extensive solidification research has significantly improved our understanding of crystal growth, but made little progress in understanding of nucleation beyond the classical nucleation theory (CNT). Due to the fact that it is almost impossible to observe the nucleus in metallic, molecular or ionic systems, most of our current knowledge of nucleation comes from simulation of homogeneous nucleation in very simple systems, while the limited experimental work is exclusively on heterogeneous nucleation in more complex systems. However, the recent advances in high resolution electron microscopy and advanced computational techniques have provided us with great opportunities to investigate heterogeneous nucleation at the atomic level. With the financial support from the EPSRC (Engineering and Physical Sciences Research Council, UK) under the LiME grant, we have made good progress in understanding heterogeneous nucleation during solidification of metallic materials. In this lecture I will provide an overview of our latest progress in understanding heterogeneous nucleation achieved in the last few years. This will include:
- Pre-nucleation phenomenon: substrate induced atomic ordering at the liquid/substrate interface. We found that a crystalline substrate can induce substantial ordering at the liquid/substrate interface at temperatures above the liquidus. We also found that pre-nucleation is promoted by reducing the misfit between the substrate and the nucleating phase.
- An epitaxial nucleation model: heterogeneous nucleation on potent substrates (i.e., with small lattice mismatch with the nucleating phase) occurs via layer-by-layer growth. Due to the presence of pre-nucleation, there is neither energy barrier nor hysteresis for epitaxial nucleation.
- Manipulation of heterogeneous nucleation: adsorption of solute/impurity atoms at the liquid/substrate interface to either enhance or poison heterogeneous nucleation. A solute (or impurity) element may segregate to the liquid/substrate interface through adsorption, forming a 2D compound or 2D solution, which in turn can either promote nucleation by reducing the lattice misfit or poisoning by increasing the lattice misfit or denying pre-nucleation.
- Competition between different types of substrates: potency vs efficiency for grain refinement. For effective grain refinement the nucleating particles need to be potent (smaller misfit), of sufficient number density, of suitable size and size distribution. Potent particles with insufficient number density may lead to coarsening, while a sufficient number of particles with large misfit may lead to significant grain refinement if there are no other particles present in the liquid.
- Classical nucleation theory (CNT) vs non-classical nucleation theory (NCNT).