ICAMS / Interdisciplinary Centre for Advanced Materials Simulation


Shedding light on the atomistic mechanisms of polymorph selection during crystal nucleation in metal alloys

Date: 22.08.2019
Place: XXVIII International Materials Research Congress, Cancún, Mexico

Grisell Díaz Leines
Jutta Rogal

Fundamental knowledge of the mechanism and the principles of polymorph selection during crystal nucleation is essential for the synthesis and control of nanomaterials with targeted properties. However, understanding the atomistic mechanism of crystallization poses a major challenge as many materials exhibit complex transitions with multiple steps, forming polymorphic structures. Here, we employ an advanced atomistic simulation method, transition path sampling (TPS), to investigate different aspects of the formation mechanism of crystal clusters during homogenous and heterogeneous nucleation in metal alloys. The statistical path ensembles obtained from TPS allow for the evaluation of kinetic and thermodynamic properties such as the free energy and the rate constant, as well as for a direct analysis of the structural composition of the growing clusters and the reaction coordinates of the process. As a first example, we investigate the nucleation pathway in unary Ni and show that the system exhibits a nonclassical mechanism, where a pre-structured liquid region is initially formed followed by the emergence of fcc-hcp crystallites embedded within the core of the growing cluster. This pre-structured cloud plays a key role in the description of the crystal clusters and the interfacial free energy, acting as a precursor of the crystallization and predetermining the polymorph selected. As a further step, we discuss how selected crystalline seeds influence the polymorphic outcome and efficiency of the nucleation in Ni. Here, fcc and hcp seeds promote the formation of pre-structured liquid clusters, enhancing the nucleation rate and determining the structural composition of the nuclei. Finally, in binary Ni3Al alloys, the nucleation pathway exhibits increased complexity due to the strong competition between chemical order and structural order of fcc and bcc phases. Here, we find that a typical single order parameter, the size of the largest cluster, is not sufficient to characterize the nucleation and a multi-dimensional description is required. Our results provide novel atomistic insight into the process of homogeneous and seeded heterogenous nucleation in metal alloys and shed light on the principles of polymorph selection in the early stages of the formation of crystalline nanoclusters.

« back