Additive Manufacturing Modeling and Simulation: Microstructure, Mechanics, and Process: AM Modeling - On-Demand Oral Presentations
Sponsored by: TMS Computational Materials Science and Engineering Committee
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
Friday 8:00 AM
October 22, 2021
Room: On-Demand Room 1
Location: MS&T On Demand
Session Chair: Jing Zhang, Indiana University - Purdue University Indianpolis
Grain Refinement and Mechanical Properties for AISI304 Stainless Steel Single-tracks by Laser Melting Deposition: Mathematical Modelling versus Experimental Results: Muhammad Arif Mahmood1; Andrei C. Popescu1; Mihai Oane1; Diana Chioibasu1; Gianina Popescu-Pelin1; Carmen Ristoscu1; Ion N. Mihailescu1; 1National Institute for Laser, Plasma and Radiation Physics (INFLPR)
A new mathematical model to calculate the number of grains and their average size inside a single printed layer via the laser melting deposition (LMD) process is proposed. The printed layer's thermal history concerning the moving laser beam and co-axial addition of powder debits was analyzed and used to calculate the thermal stress and strain rate. The average grain size within the printed layer was calculated using the Johnson-Mehl-Avrami-Kolmogorov (JMAK) model. The mechanical properties, including ultimate tensile strength, yield strength, and hardness, were estimated using the average grain size. For single depositions of AISI 304 stainless steel powder debits on a steel substrate, dedicated experiments were performed to verify the model’s trustworthiness. Scanning electron microscopy was used to quantify the number and size of grains. Vickers hardness tests were conducted to confirm the mechanical. The model was able to estimate results with 10-15 % mean absolute deviations.
Numerical Simulations of Fracture Tests of Uncharged and Hydrogen-charged Bend Specimens of Additively Manufactured 304 Stainless Steel Using Nodal Release Method and Cohesive Zone Model: Shengjia Wu1; Shin-Jang Sung1; Jwo Pan1; Paul Korinko2; 1University of Michigan; 2Savannah River National Laboratory
Continuum modeling of fracture behavior is important for additively manufactured materials. Continuum finite element analyses are used to model crack extensions in uncharged and hydrogen-charged bend specimens of laser directed energy additively manufactured 304 stainless steel using the nodal release method and cohesive zone model. The maximum opening stresses ahead of crack tips and separation work rates as functions of the crack extension are first obtained by the nodal release method as references for the varying cohesive strength and cohesive energy. The load-displacement-crack extension relations are then obtained using the cohesive zone model with the varying cohesive parameters by fitting the experimental data for the uncharged and hydrogen-charged specimens. The computational load-displacement-crack extension relations obtained from the finite element analyses using the cohesive zone model with the varying cohesive parameters can fit very well with the experimental data.
Ductile Fracture of Ti-6Al-4V Made by Powder Bed Fusion Additive Manufacturing: Allison Beese1; Alexander Wilson-Heid1; 1Pennsylvania State University
Additively manufactured Ti-6Al-4V manufactured via laser powder bed fusion (L-PBF) was subjected mechanical tests to investigate the impacts of microstructure on deformation and fracture behavior. To study the material’s plasticity and fracture behavior under multiaxial loads, tests were performed over a wide range of tension and tension/shear stress states, and in two material directions. A plasticity model was calibrated and validated, which enabled the study of ductile fracture. An experimental/computational approach was used to identify the loading history of stress state and strain accumulation to the point of fracture under the stress states studied. This presentation will describe the fracture models used to capture these data as well as the impact of defects on fracture of this alloy.
Thermal History of LPBF Components Towards Predicting As-built Material Properties: Martin Verhülsdonk1; Simon Vervoort1; Mustafa Megahed2; 1Fraunhofer Institute for Laser Technology ILT; 2ESI Group
LPBF as-built material properties, residual stress and component shape are dependent on the thermal history. Design and optimization of the deposited material thus far depends on experimental DOE’s – a systematic approach for industrial applications based on model informed optimization is missing. The numerical challenge is to predict the thermal history of a printed component using the machine build file. In this work, the conduction equation is reformulated to enable accelerated simulations. The laser position and operating conditions are read from the build file. The laser trajectory throughout the component is resolved providing required temperatures and gradients required for thermodynamic analysis. A demonstration build exhibiting both hot and cold regions is used to induce different metallurgical responses of the deposited IN718. Thermocouples are used to validate numerical predictions. Phase transformations are simulated and compared with component characterization confirming the accuracy of the overall process.
Defect Prediction thru Part-scale Simulation: Shawn Hinnebusch1; Florian Dugast1; Alaa Olleak1; Albert To1; 1University of Pittsburgh
Defect formation within parts is one of the leading issues with additive manufacturing (AM) during the laser powder bed fusion (L-PBF) process. AM parts can fail pre-maturely making it difficult to qualify parts for critical components. For this reason, AM parts are not used to their full potential. A remedy to this issue is using simulations to predict defects before printing. Microscale simulations are computationally expensive because the large number of elements make it impractical to simulate a full-scale simulation. This work makes use of a GPU accelerated process simulation to quickly complete a transient heat transfer model. A layer-by-layer approach is implemented where the results are used to locate potentially problematic areas within the print. In those areas, a microscale simulation is completed to help quantify the defects in those regions. This approach significantly reduces the computational time while still capturing the microscale details required to ensure part quality.
Inherent Strain Method for Residual Stress Prediction in Ferritic-austenitic Steel Structure Fabricated by Directed Energy Deposition: Zhengtong Shan1; Minh Tien Tran1; Huai Wang2; Sun-Kwang Hwang3; Dong-Kyu Kim1; 1University of Ulsan; 2Chinese Academy of Sciences; 3Korea Institute of Industrial Technology
The present study investigated the inherent strain method for residual stress prediction in ferritic-austenitic steel structure fabricated by directed energy deposition. A coupled thermo-mechanical analysis of a micro-scale model was applied to obtain the inherent strains for analysis of residual stress. The predicted residual stress is in good agreement with experiment results measured by the contour method and neutron diffraction. Furthermore, the effect of dissimilar structures and scanning strategy on the residual stress is discussed in this study.
Deep Learning Prediction of Stress Fields in Additively Manufactured Metals with Intricate Defect Networks: Brendan Croom1; Michael Berkson1; Bobby Mueller1; Michael Presley1; Steven Storck1; 1JHU Applied Physics Laboratory
In context of the universal presence of defects in additively manufactured (AM) metals, efficient computational tools are required to rapidly screen AM microstructures for mechanical integrity. To this end, a deep learning approach is used to predict the elastic stress fields in images of defect-containing metal microstructures. A large dataset consisting of the stress response of 100,000 random microstructure images is generated using high-resolution Fast Fourier Transform-based finite element (FFT-FE) calculations, which is then used to train a modified U-Net style convolutional neural network (CNN) model. The trained U-Net model more accurately predicted the stress response compared to previous CNN architectures, and exceeded the accuracy of low-resolution FFT-FE calculations. The model was applied to images of real AM microstructures with severe lack of fusion defects, and predicted a strong linear increase of maximum stress as a function of pore fraction.
Process Maps and Models For Highly Filled Polymers In Powder Fused Filament Fabrication (PF3) 3D Printing: Kameswara Pavan Kuma Ajjarapu1; Roshan Mishra1; Ji-Hae Kim1; Kunal Kate1; 1University of Louisville
Powder fused filament fabrication (PF3) 3D printing utilizes powder-filled polymer filaments, while combining Fused filament fabrication (FFF) and sintering processes to fabricate complex metallic and ceramic structures. As PF3 filaments tend to have >85 wt.% powder content within them, they often exhibit low filament strength and flow instabilities that result in print failures. In this work, unfilled filaments and highly filled (90wt.%) filaments were printed via PF3 at different extrusion temperatures and print speeds in order to model the rheology, flow rate and mechanical properties required to print parts with a relative density >90%. Additionally, for the same printed parts, process maps were generated to identify viscosity and shear rate regions. Further, rheological, thermal and mechanical characterization was performed to identify how material properties of the filaments affect printability. It is expected that such a process model will facilitate identification of optimum print parameters for a given material system.
Modeling Collapse Behavior in Large-scale Thermoset Additive Manufacturing: Stian Romberg1; Chris Hershey2; John Lindahl2; Abrian Abir3; Michael DeVinney4; Chad Duty1; Vlastimil Kunc2; Brett Compton3; 1University of Tennessee, Knoxville; Oak Ridge National Laboratory; 2Oak Ridge National Laboratory; 3University of Tennessee, Knoxville; 4
Over the last several years, 3D printing of thermoset polymer resins has progressed rapidly. Thermosets offer desirable properties and excellent compatibility with fibers. Additionally, they can be deposited at room temperature, avoiding the residual stresses and poor interlayer bonding caused by the large thermal gradients in thermoplastic material extrusion additive manufacturing (AM). However, pursuits to scale up thermosets have highlighted issues with print stability. This presentation will describe efforts to quantify deformation mechanisms in large-scale printed thermosets and to link measurable rheological properties to collapse in tall, thin printed walls using self-weight buckling and yielding models. Findings reveal that the observed buckling instability is governed by the recovered storage modulus of the material after a period of high shear stress representative of that experienced in the deposition nozzle. The talk will conclude with discussion of how these results can inform material design and print parameter selection for large-scale thermoset AM.
Interfacial Properties in 3D Printed Stainless Steel Coated with Epoxy: Xuehui Yang1; Sugrim Sagar1; Tejesh Dube1; Alan Jones1; Jing Zhang1; 1Indiana University – Purdue University Indianapolis
Epoxy coating is one of the most common corrosion mitigation methods to protect metal substrates from water, dissolved oxygen, and corrosive ions. However, there is limited research conducted on the coatings on the additively manufactured (AM’d) metals. In this work, the epoxy coating was applied on the surface of AM’d 316L SS samples. To explore the interfacial properties, the bond strength was tested on the interface between the coating layer and 316L SS samples, and the corrosion properties were also explored with both coated samples and AM’s samples. To explore the fracture mechanism of the interfacial bond, a molecular dynamics (MD) model was built to simulate the interface between the polymer and metal. The simulated pull-off test was conducted to evaluate the mechanical properties and failure mechanics on the interface. This study provides important information for extending the AM metals in applications where corrosion resistance is required.
Mechanical Properties of Ceramic Core with SiO2-Na2O-Al2O3 Ternary Binder System: Hyunhee Choi1; Bong-Gu Kim2; Eun-Hee Kim1; Junseong Kim2; Seong-Hwa Jeong2; Seung-Cheol Yang2; Yeon-Gil Jung2; 1Changwon National University; 2Department of Materials Convergence and System Engineering of Changwon National University
The organic-inorganic binder conversion process was introduced to improve the strength of ceramic core in a casting process. As of now, the binary inorganic binder system of Na2O-SiO2 has been applied in the conversion process. However, there are limitations in the thickness and shape of cast products, owing to the working temperature of the binary binder system. Therefore, in this study, a ternary inorganic binder system composed of SiO2-Na2O-Al2O3 was introduced in the conversion process. As the amount of Al precursor increased in the ternary binder system, mechanical properties, such as Young's modulus, fracture toughness, and hardness, were improved, which showed similar values and trends to those calculated through molecular dynamics. Finally, by applying the ternary binder system to the fabrication process of the core, the strength of the core was improved and the casting possibility was confirmed in high temperatures.
Preparation of Ceramic Green Body with Uniform Density through Living Properties of Cycloaliphatic Epoxy Resins in DLP(Digital Light Processing) 3D Printing of Ceramics: Hye-Yeong Park1; SeungHwa Jeong2; Haeun Kim2; DongHyun Kim2; Janghyeok Pyeon2; SeungCheol Yang1; Yeon-Gil Jung1; 1Changwon National University; 2Department of Materials Convergence and System Engineering of Changwon National University
In a photo-polymerization based ceramic 3D printing, the opaque slurry reduces the penetration depth of light, resulting in a difference in crosslinking density within the layer, which causes problems such as cracking and distortion when fabricated into a firing body. In this study, the cycloaliphatic epoxide has been proposed as a monomer of photo-curable slurry to fabricate the green body with a uniform crosslinking density due to the living characteristics of photo-cationic polymerization. If there is no nucleophilic substance in the cationic polymerization, the polymerization may continue to be proceeded even after light irradiation. As a result, the ceramic slurry produced using the cycloaliphatic epoxide and silica underwent an additional polymerization by heating at 150°C after the photo-polymerization, showing a uniform the crosslinking density and improving the green strength. In addition, it was possible to produce a sound fired body without any change in a shape after firing at 1000°C.