Precipitates in nickel-based superalloys: insights from atomistic simulation, by Peter Brommer

This seminar is by Peter Brommer of Warwick University, organised by Prakash Srirangam.
Nickel-based superalloys are remarkable materials. Due to their high strength even at elevated temperatures they are used for turbine blades in jet engines. Their properties are intrinsically linked to their microstructure, where the ordered γ’ phase forms precipitates in the γ phase solid solution.
Atomistic simulations can offer key insights into their performance. In this talk, I will showcase two ongoing studies into different aspects of microstructures in nickel-aluminium alloy systems, demonstrating the capabilities and limitations of this simulation method.
Firstly, precipitates form during heat treatment on a timescale inaccessible to direct molecular dynamics simulations. Kinetic Monte Carlo methods, where the system is moved from one atomic configuration to the next, offers a pathway to accelerate simulated time. This could allow a determination of precipitate growth rates dependent on temperature and chemical composition, particularly for novel processing methods such as additive manufacturing, but this requires reliable values for the barrier energies separating distinct atomic configurations. While we show that existing interatomic potentials for the nickel-aluminium system cannot provide those, modern machine-learned interatomic potentials could offer a way out - but they come with their own challenges.
Secondly, the extraordinary strength of these materials is primarily caused by the interaction of dislocations with the nanoscale precipitates. We fit the parameters of the equation of motion of an edge dislocation in face-centred cubic nickel representing an idealised γ phase using Differential Evolution Monte Carlo within a Bayesian framework. This procedure allows us to propagate the uncertainty through the model to the model predictions: dislocation position and velocity. This serves as a first step towards building a more comprehensive surrogate model that can describe the deformation behaviour of Ni-based superalloys, where the parameterisation approach employed here can offer a quantitative measure of the effect of precipitates on dislocation dynamics in these materials.
Peter Brommer is an Associate Professor in the School of Engineering, University of Warwick, UK. He gained his PhD in theoretical physics from the University of Stuttgart, Germany. After postdoctoral posts in Stuttgart, Montréal (Québec, Canada) and Warwick Physics, he joined the Warwick Centre for Predictive Modelling in 2014. His main area of research is computational materials science, with a particular focus on atomistic simulation of metals and alloys. His passion is creating and improving models that bridge length and time scales while quantifying their uncertainties, to permit linking from quantum mechanics up to experimentally accessible systems.