Additive Manufacturing Modeling, Simulation, and Machine Learning: Microstructure, Mechanics, and Process: Microstructures & Defects II
Program Organizers: Jing Zhang, Purdue University in Indianapolis; Brandon McWilliams, US Army Research Laboratory; Li Ma, Johns Hopkins University Applied Physics Laboratory; Yeon-Gil Jung, Korea Institute of Ceramic Engineering & Technology

Wednesday 8:00 AM
October 12, 2022
Room: 303
Location: David L. Lawrence Convention Center

Session Chair: Jing Zhang, Indiana University - Purdue University Indianapolis; Yeon-Gil Jung, Changwon National University; Li Ma, Johns Hopkins University Applied Physics Laboratory; Brandon McWilliams, CCDC Army Research Laboratory

8:00 AM
Energy and Microstructural Evolution of In-situ Alloyed Cu-4at% Cr -2 at% Nb via Laser Powder Bed Fusion: David Scannapieco¹; David Ellis²; John Lewandowski¹; ¹Case Western Reserve University; ²NASA Glenn Research Center
    Literature suggests that the energy needed to in-situ alloy material via additive manufacturing is higher per unit volume than their pre-alloyed counterparts, often despite the in-situ formation of thermodynamically favorable phases. This work, via experiments on in-situ alloyed GRCop-42 (Cu-4 at% Cr-2 at% Nb), will explore the energy needs of an elemental powder blend during in-situ alloying via laser powder bed fusion (LPBF). Empirically derived models will explore the influence of elemental thermophysical properties, net energy of in-situ reactions, and powder characteristics on the resulting LPBF energy demands, weld pool size and shape, and porosity evolution. Additionally, the influence of energy input during printing on the in-situ reaction to form Cr2Nb will be explored. Evolution of Cr2Nb dispersoid and copper grain shape, size, and texture can be related to the temperature reached, cooling rate, and time-as-melt during in-situ alloying via LPBF.

8:20 AM
Microstructure Predictions in Additive Manufacturing from Analytical Solidification Models – A Critical Assessment of Simplifying Assumptions: Jonah Klemm-Toole¹; Charles Smith¹; Olivia DeNonno¹; Matthew Schreiber¹; Luc Hagen¹; Gwilym Couch¹; Zhenzhen Yu¹; Kip Findley¹; John Speer¹; Joy Gockel¹; Amy Clarke¹; Tony Petrella¹; Craig Brice¹; ¹Colorado School of Mines
    Prediction of the as-solidified microstructure in additive manufacturing (AM) is a key capability that is needed to develop the next generation of advanced materials and processing technologies. Analytical models offer easy to use expressions that can be quickly solved enabling the rapid assessment of material and process variables. However, many analytical solidification models were developed for highly simplified conditions. In this presentation, we discuss how these models can be extended to complex alloys and higher solidification rates relevant to AM. Using austenitic stainless steels as an illustrative example, we demonstrate how common simplifying assumptions can affect predictions of solidification mode, primary and secondary dendrite arm spacing, dendrite growth morphology, and primary solidifying phase under processing conditions representative of several AM processes. The insights provided in this talk are intended to spark rich discussions on how the fundamentals of solidification can be analytically modeled to advance our understanding of AM.

8:40 AM
Development of Rapid Solidification Model for Additive Manufacturing and Application to Al-Si Alloys: Minho Yun¹; In-Ho Jung¹; ¹Seoul National University
    Metal additive manufacturing has numerous advantages in industries, however due to its complexity, estimating product’s microstructure is a challenging, but yet essential task to solve. Therefore, current study focuses on developing rapid solidification model that could be applied to selective laser melting. Various models for solidification were adopted to advance the model. Using process parameters, primary dendrite arm spacing (PDAS), primary cell fraction, and solute profile were predicted. Single track experiments with various process condition using Al-12Si binary alloy were held to compare and verify the presented model. Authentication was also done by using empirical data from single track experiment from other studies. Utilizing single semi-empirical parameter, both PDAS and microsegregation model showed a great agreement with experimental data. Deviation between experiment and model was elucidated by abnormal property of Al-Si alloys. Furthermore, PDAS estimation for various concentration and predicting modified eutectic point were done for application.

9:00 AM
Modeling the Solidification Cracking Susceptibility of Additively Manufactured Alloys: Noah Sargent¹; Soumya Sridar¹; Richard Otis²; Wei Xiong¹; ¹University of Pittsburgh; ²Jet Propulsion Laboratory, California Institute of Technology
    Deviation from thermodynamic equilibrium during rapid solidification in additive manufacturing processes causes microsegregation, depression of the solidus temperature, and increased susceptibility to solidification cracking. Modeling the combined effect of kinetics and thermodynamics on solidification in complex multi-component systems is a challenge because part-scale thermal models for the additive manufacturing process and multi-component solidification models must be linked together. This work covers recent work on modeling the effect of thermal history on solidification in both laser powder bed fusion and wire arc additive manufacturing processes. Thermal modeling and experiments are integrated into a CALPHAD-based ICME framework to predict the impact of processing parameters, build geometry, and composition on solidification cracking susceptibility. Modeling results show that increasing the energy density reduces the solidification cracking susceptibility of stainless steel 316L made with laser powder bed fusion, and location-specific thermal history impacts the solidification cracking susceptibility in the wire arc additive manufacturing process.

9:20 AM
Sparse Sampling for 3D Electron Backscatter Diffraction: Zachary Varley¹; Gregory Rohrer¹; Marc De Graef¹; ¹Carnegie Mellon University
    Electron backscatter diffraction (EBSD) is a popular microstructure analysis technique due to the direct, relatively high-resolution, measurement of phase and orientation. Three-dimensional EBSD (3D-EBSD) extends this microstructure analysis to a volume of material, measured in 2D sections, with a commensurate penalty in data acquisition time. The present work proposes a sampling routine to leverage the organized granular structure of orientation data present in microstructures, both within individual serial sections, and across consecutive ones. By using online machine learning to avoid redundant measurements, and then infilling unmeasured data during post-processing, significant theoretical savings are realized without offline training. User parameters allow tuning of the tradeoff between reconstruction accuracy and sampling speed. By coupling this sampling approach with real-time Kikuchi pattern indexing the authors aim to create a robust 3D-EBSD sampling technique which improves the Pareto frontier in the tradeoff between scan time and sample volume.

9:40 AM
Printability and Failure Susceptibility of Different Powder Layer Thicknesses in Laser Powder Bed Fusion: Tayler Sundermann¹; David Shoukr¹; Peter Morcos¹; Raymundo Arroyave¹; Alaa Elwany¹; Ibrahim Karaman¹; ¹Texas A&M University
    Superalloy 718 has been extensively built in Laser Powder Bed Fusion and specifically within this group the development of an efficient process optimization framework that would cover the entire scanning speed (v)-laser power (P)-hatch spacing-layer thickness parameter space. During the exploration of this framework for layer heights beyond 60痠 envelope; standard performance characteristics such as % porosity was observed to be disconnected from mechanical performance. Upon further investigation, in a printability map comprised of speed, laser power, hatch spacing, and layer thickness above 60痠 there is a decreasing window of printability which is susceptible to simultaneous lack of fusion and keyholing porosity defects. To print effectively in this region could present an increase in throughput rate but is accompanied by additional challenges. This presentation will show the role of powder layer thickness on the optimum printability region and consideration of challenges to increasing layer heights.