6th World Congress on Integrated Computational Materials Engineering (ICME 2022): Applications: Advanced Manufacturing – Additive Manufacturing I
Program Organizers: William Joost; Kester Clarke, Los Alamos National Laboratory; Danielle Cote, Worcester Polytechnic Institute; Javier Llorca, IMDEA Materials Institute & Technical University of Madrid; Heather Murdoch, U.S. Army Research Laboratory; Satyam Sahay, John Deere; Michael Sangid, Purdue University
Monday 10:30 AM
April 25, 2022
Room: Regency Ballroom AB
Location: Hyatt Regency Lake Tahoe
Session Chair: Kester Clarke, Los Alamos National Laboratory
10:30 AM Invited
Predicting As-Built Additively Manufactured Microstructures and Residual Stress Using the CAFE Model: Kirubel Teferra1; Lukasz Kuna; David Rowenhorst1; 1Naval Research Laboratory
As a result of the high dimensional space associated with additively manufacturing (AM) build conditions, it is advantageous to have predictive models capable of mapping laser parameters to a material’s microstructural state. In prior work, an implementation of the cellular automata finite element (CAFE) model optimized for AM processing has been developed and validated for bulk material regions for powder bed fusion 316L in terms of polycrystalline grain morphology and crystallographic texture. In the present work, this model is extended to consider the effects of free surface and corner conditions commonly found in AM builds of struts or thin walls, where contouring scan strategies are employed to eliminate deleterious artifacts. Additionally, the residual stress in the as-built microstructures is calculated by coupling the CAFE model to a finite element model computing the thermally driven stress fields. This presentation details the implementation and demonstrates its validation.
11:00 AM
Exascale Cellular Automata for Simulating Grain Structures in Additive Manufacturing: Sam Reeve; Matthew Rolchigo1; Jim Belak1; 1Lawrence Livermore National Laboratory
Simulating the entire process of metallic additive manufacturing (AM) is making significant strides towards helping understand the structures and concomitant properties in experimental AM parts and, eventually, increased use in application. Cellular automata (CA) is a crucial step in understanding process-microstructure-property relations in AM, where we have developed the ExaCA application with a sparse representation of the time-temperature history from process models to drive simulation of grain-scale microstructures that can be used to determine constitutive properties. We focus specifically on the computational aspects of running and improving ExaCA, particularly on current and future exascale architectures, including multi-core CPUs and GPUs. Important points of emphasis are writing code in a performance portable manner with the Kokkos programming model and considering optimizations which fit properly within the larger ExaAM (Exascale Computing Project for AM) AM build workflow as described above.
11:20 AM
Multiscale Modeling of Microstructure Formation in Directed Energy Deposition for Nickel-Based Superalloys: Lang Yuan1; Siyeong Ju2; Shenyan Huang2; Yiming Zhang2; Yang Jiao2; Chen Shen2; Hyeyun Song3; Luke Mohr3; Lee Kerwin3; Jason Parolini4; Changjie Sun2; Alex Kitt3; 1University of South Carolina; 2GE Research; 3EWI; 4GE Power
Directed Energy Deposition (DED) is one of the enabling metal additive manufacturing technologies to print functionally graded materials. In DED, the complex thermal history determines the as-built microstructures and defects. In this study, a physics-based process model was developed and validated against in-situ process monitoring data, predicting the materials deposition process with transient thermal profiles. The temperature profiles were then coupled into a highly parallel solidification microstructure model to predict the grain structures at both the meso- and micro-scale. The multiscale model, taking inputs from theoretical solutions from solidification theory and thermodynamic modeling, was applied to study the nucleation and grain growth in single tracks and thin walls for two nickel-based superalloys, IN718 and Rene 80. The predictions are compared against materials characterizations. Its application to assist the materials design and process development for functionally graded materials in DED were discussed and demonstrated, illustrating a powder tool implementing ICME approach.
11:40 AM
Prediction of the Columnar-to-Equiaxed Transition During Additive Manufacturing of Concentrated Multicomponent Alloys: Christopher Hareland1; Peter Voorhees1; 1Northwestern University
The properties of additively manufactured components are governed by the solidification microstructures that naturally arise during processing. One example is the columnar-to-equiaxed transition (CET), in which columnar grains suddenly give way to an equiaxed microstructure. Existing models of the CET are limited by overly simplistic models of dendritic growth that neglect effects that arise when working with concentrated, multicomponent alloys, such as diffusional interactions between solute species. Here, we derive a general expression for the partition coefficient and perform a marginal stability analysis to determine the tip radius for dendritic solidification of a multicomponent alloy without making the common simplifying assumption of a dilute-ideal solution. By employing CALPHAD free energies of the liquid and solid, this general model of dendritic growth can be used in a CET model to more accurately calculate the processing conditions required to induce the CET during additive manufacturing of modern, technologically relevant alloys.
12:00 PM
Heat Source Sizing for FEA of NAB Using Wire-Fed AM: Chris Jasien1; Charles Fisher; 1Naval Surface Warfare Center - Carderock
As the landscape of large-scale additive manufacturing (AM) quickly advances, integrating materials simulation techniques into the design framework is essential to accelerating development. Simulation enables reduction in process design time and residual stress prediction, plus optimization of programmed deposition paths as well. In this research, finite element analysis (FEA) modeling of both an arc and laser heat source were performed for Nickel-Aluminum-Bronze (NAB) wire deposition. Single-pass scenarios using a range of Goldak and Gaussian heat source dimensions were examined and compared to cross-sectional images of physical builds for verification and validation of the predicted melt pools. Results informed the selection of heat source dimensions for specific thermal scenarios. Additional multi-pass scenarios investigated thermal cycle effects on multi-layer parts, which is essential for understanding microstructural development within large-scale builds. All of the models provided valuable information for optimized processing conditions for large-scale builds of NAB material using wire-based AM techniques.