The prediction of solidification microstructures associated with additive manufacture of metallic components is fundamental in the identification scanning strategies, process parameters and subsequent heat treatments for optimised component properties. Interactions between the powder particles and the laser heat source result in complex thermal fields in and around the metal melt pool, which will influence the spatial distribution of chemical species as well as solid-state precipitation reactions. This paper demonstrates that a multi-component, multi-phase precipitation model can successfully predict the observed precipitation kinetics in Inconel 625, capturing the anomalous precipitation behaviour exhibited in additively manufactured components. A computer coupling of phase diagrams and thermochemistry (CALPHAD)-based approach captures the impact of dendritic segregation of alloying elements upon precipitation behaviour. The model was successful in capturing the precipitation kinetics during annealing considering the Nb-rich and Nb-depleted regions that are formed during additive manufacturing.

This paper presents a computational framework to study the differences in process-induced microvoid and precipitate distributions during selective laser melting (SLM) of two nickel-based superalloys representative of low (IN718) and high (CM247LC) volume fraction precipitate-strengthened alloys. Simulations indicate that CM247LC has a higher propensity to form process-induced microvoids than IN718. Particle sintering is predicted to be strongly influenced by the powder size distribution. For deposition thickness of approximately 40 μm, thermal gradients during cooling are predicted to be larger for CM247LC than IN718 and consequently expect the development of larger residual stresses for a high volume fraction γ ′ alloy. A coupled mean field/finite-element approach has been used to predict the precipitate distributions across a simple rectangular build and during a subsequent hot isostatic pressing (HIP) cycle. Unimodal and multi-modal particle distributions are predicted for IN718 and CM247LC at the end of the SLM, respectively. A higher volume fraction of γ ′ is predicted for CM247LC at the end of the SLM process. During HIP, simulations indicate a dramatic increase in the γ ′ volume fraction in CM247LC, which can result in a reduction in stress relaxation and lead to a ductility drop.

Materials modeling technologies are fundamental to explore, understand, and ultimately predict materials behavior. They are essential to solve challenges posed by the need to reduce human impact on the environment. Modeling and simulation of materials behavior have been recognized over the years as fundamental as an asset in industrial R & D, guiding the decision-making process regarding the design or optimization of new products and manufacturing processes. At the same time, it reduces product cost and development time. However, highlighting the revenue brought by using such tools is not trivial, especially because they mainly affect the complex activities such as the innovation process, whose return only becomes available in the long run and it is difficult to measure. This means that the materials modeling field is often overlooked in an industry setting, where it is not integrated in the company workflow. In some cases, modeling provides the potential to capture tacit knowledge preventing the loss of capability in an aging specialist community, that why its industrial integration is important. This paper explores the reason behind this dichotomy, presenting first what it is intended for the modeling process, and the main types used in materials application. The current industrial adoption is reviewed by outlining success stories, economic impact, business uptake, and barriers. Past and current approaches and strategies are also presented and discussed. In prospective, materials modeling plays a key role in developing material-centric industry for sustainable economy, providing physical understating (physics-based models) and fast approaches (data-driven solutions). Digitalization is the mean for the green economy and it needs to push for a more integration at the core of the business of materials modeling.

Automated stereological methods are presented for approximating the 3D size distribution of unimodal or bimodal precipitate dispersions considering 2D and 1D measurements taken from polydisperse spherical non-penetrating particle dispersions. A method to quantify the uncertainty of the approximation as a function of the number of sampled particles is presented and demonstrated to experimental data. The derivation and verification of the analytical stereological expressions used are included. Two procedures are presented for estimating the 3D size distribution of bimodal particle populations depending upon the relative size of the two particle populations. If the particles can be characterised using micrographs of the same magnification, it is possible to estimate the volume fraction of each particle population. For cases where micrographs have been taken at different magnification, an estimate of the area fractions of the particle populations is needed to combine the datasets and allow for the approximation of the 3D size distribution. These methods are useful for use in determining the initial particle size distribution for use in modelling and determining the appropriate number of micrographs and particles to measure when characterising a precipitate dispersion.

This article elucidates the need to consider the inherent spatial transfer function (blur), of any thermographic instrument used to measure thermal fields. Infrared thermographic data were acquired from a modified, commercial, laser-based powder bed fusion printer. A validated methodology was used to correct for spatial transfer function errors in the measured thermal fields. The methodology was found to make a difference of 40% to the measured signal levels and a 174 °C difference to the calculated effective temperature. The spatial gradients in the processed thermal fields were found to increase significantly. These corrections make a significant difference to the accuracy of validation data for process and microstructure modeling. We demonstrate the need for consideration of image blur when quantifying the thermal fields in laser-based powder bed fusion in this work.

The ability to accurately predict the time evolution of precipitate size distributions is fundamental to optimising heat treatments and mechanical properties of engineering alloys. Mean-field models of the particle growth rates assume that diffusion fields between neighbouring particles are weakly coupled reducing the problem to a single particle embedded in an effective medium. This regime of behaviour is expected to be satisfied for low volume fraction alloys. However, these assumptions are not fulfilled in many applications of interest where strong interactions between precipitates holds. Correction factors are often introduced to account for the accelerated rate of diffusion caused by the overlapping of diffusion fields between neighbouring precipitates. This paper applies the Wang–Glicksman–Rajan–Voorhees (WGRV) discrete point-source/sink model to compare descriptions of competitive growth. This includes assessing correction factors to the mean-field particle growth rate derived by Ardell, Marqusee and Ross, and Svoboda and Fischer in addition to Di Nunzio’s pairwise interaction model. The WGRV model is used as a benchmark to compare different approximations of competitive growth that apply similar assumptions. This is followed by the application of the models to simulate precipitation kinetics during long term aging kinetics observed in the nickel-based superalloys IN738LC and RR1000. It is shown that the competitive growth correction factors are accurate for volume fractions of 20% and under-predict the acceleration of precipitate kinetics predicted at 40%. The WGRV model is able to capture the coarsening kinetics observed in both IN738LC and RR1000 with reasonable accuracy. The WGRV model determines particle growth rates as a function of the immediate neighbourhood and provides an improved prediction of the coarsening behaviour of tertiary particles in RR1000 in comparison to the mean-field approximation, however over-estimates the growth rate of the tertiary particles compared to experimental data.