Precipitation is accompanied by the formation and migration of heterophase interfaces. However, the detailed structures of the precipitate interfaces have rarely been characterised at the atomic scale until recently, revealing complicated but well-defined interfacial structures. This fact is not considered in classical analytical models, which assume the interface to be sharp or diffused. The lack of precise experimentally-based interfacial structures prevents accurate atomistic predictions even for the simplest cases.
To address this challenge, we combine electron microscopy characterisation with atomistic calculations. The alloy microstructure and precipitate atomic structures were characterised by high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) to exploit the atomic number difference between solute elements and the matrix. High-resolution imaging was conducted in a dual-aberration-corrected FEI Titan3 80-300 field-emission gun transmission electron microscope (FEGTEM) operated at 300 kV while preliminary investigations were carried out on a JEOL 2100F and an FEI Tecnai F20 FEGTEM operated at 200 kV. Various microscopy techniques were applied, including parameter estimation theory for atomic column position refinement, in situ heating for microstructure evolution and electron tomography for 3D structural analysis. For atomistic calculations, we used density functional theory for accurate energetics calculations as implemented in VASP. To deal with large systems containing complicated interfacial structures, we performed molecular dynamics simulations as implemented in LAMMPS, using embedded atom method and recently developed deep learning potential.
The γ’ (Ag2Al) phase in the Al-Ag alloy system has served as a textbook example for understanding phase transformations, precipitating hexagonal close-packed (HCP) crystals in the face-centred cubic (FCC) aluminium matrix. The γ’ precipitates display fully coherent interfaces at their broad facets and semicoherent interfaces at their edges. By bridging advanced microstructure characterisation and atomistic simulations, we determined the exact locations and core structures of interfacial dislocations. In particular, we discovered a new FCC/HCP interfacial structure that displays a unique combination of Shockley partial, Lomer-Cottrell and Hirth dislocations that evolve from the known interfacial structure purely composed by Shockley partial dislocations, as shown in Fig.1. Our findings show that the FCC-HCP transformation is more complex than hitherto considered, due to the interplay between structure and composition confined at interfaces. Despite its textbook status, the Al-Ag system has not revealed all its phase transformation secrets. For instance, we recently discovered a new phase ζ (AgAl) in the binary system , as shown in Fig.2.
Using this combined approach of advanced imaging and atomistic simulation, we also studied the precipitate-matrix interfaces in various other simple systems, including Al-Cu[3,4], Al-Au, Al-Cu-Au, Al-Cu-Ag and Al-Cu-In-Sb alloys. Our results demonstrate that DFT calculations can predict reliably the segregation behaviour of solute at precipitate interfaces, at least when the structural configurations of the segregation are known. This corroborates earlier work on the segregation at precipitate interfaces in aluminium and implies that first-principles computations will become increasingly important in the quantitative evaluation of the structural characteristics of precipitate phases when used in concert with atomic-scale experimental characterisation techniques.
Fig.1. The coupling between chemical distributions and dislocations at semicoherent Al-γ’ FCC/HCP interfaces.
Fig.2. (a) The atomic structure of ζ phase and in-situ evolution, showing alternating Al- and Ag-enriched bi-layers on close-packed planes. (b-e) Microstructure evolution of ε-ζ-γ′ phase transformation as imaged during in situ annealing at 200°C.
 Zhang, Z., Rosalie, J.M., Medhekar, N.V. and Bourgeois, L., Acta Materialia 174 (2019) p. 116-130.
 Zhang, Z., Bourgeois, L., Rosalie, J.M. and Medhekar, N.V., Acta Materialia 132 (2017) p. 525-537.
 Bourgeois, L., Dwyer, C., Weyland, M., Nie, J. F., & Muddle, B. C., Acta Materialia 59 (2011) p. 7043-7050.
 Bourgeois, L., Medhekar, N.V., Smith, A.E., Weyland, M., Nie, J.F. and Dwyer, C. Physical review letters 111 (2013) p. 046102.
 Bourgeois, L., Zhang, Z., Li, J., & Medhekar, N. V., Acta Materialia 105 (2016) p. 284-293.
 Chen, Y., et al. Acta Materialia 125 (2017) p. 340-350.
 Rosalie, J. M., Bourgeois, L., & Muddle, B. C. Acta Materialia 59 (2011) p. 7168-7176.
 Zhang, Y., Zhang, Z., Medhekar, N.V. and Bourgeois, L. Acta Materialia 141 (2017) p. 341-351.
 The authors acknowledge funding from the Australian Research Council (LE0454166, LE110100223, DP150100558), computational support from the Monash Sun Grid cluster, the National Computing Infrastructure and Pawsey Supercomputing Centre funded by the Australian Government, and the use of facilities within the Monash Centre for Electron Microscopy.