Additive Manufacturing Fatigue and Fracture: Developing Predictive Capabilities: Surface Roughness and Porosity Effects
Sponsored by: TMS Structural Materials Division, TMS: Additive Manufacturing Committee, TMS: Mechanical Behavior of Materials Committee
Program Organizers: Nik Hrabe, National Institute of Standards and Technology; John Lewandowski, Case Western Reserve University; Nima Shamsaei, Auburn University; Mohsen Seifi, ASTM International/Case Western Reserve University; Steve Daniewicz, University of Alabama

Tuesday 2:30 PM
March 1, 2022
Room: 258B
Location: Anaheim Convention Center

Session Chair: Nik Hrabe, National Institute of Standards and Technology (NIST); Jake Benzing, National Institute of Standards and Technology (NIST)

2:30 PM  Invited
Additive Manufacturing Surface Roughness Formation, Characterization, and Influence on Fatigue Performance: Joy Gockel1; Rachel Evans2; Edwin Glaubitz1; Anna Dunn2; Wesley Eidt2; 1Colorado School of Mines; 2Wright State University
    Additive manufacturing provides the ability to fabricate complex geometries. The inherent surface roughness that is produced is difficult to post-process and detrimental to the mechanical performance. The surface formation is influenced by the manufacturing processing parameters and the local geometry. Surface characterization methods and the relationships connecting the processing parameters and melt pool geometry to the surface roughness in laser powder bed fusion will be discussed. The effect of the surface roughness on the fatigue performance is determined using standard and custom test coupon geometries for axial and vibration-based bending fatigue. Results suggest that surface roughness is a significant factor when the rough surface is at high stress locations. However, competing mechanisms such as porosity, microstructural features and component shape also influence the failure in structurally relevant geometries. Understanding the surface roughness provides guidance for design rules and post-processing development to reduce the impact of surface roughness on fatigue failure.

3:00 PM  
The Role of As-built Surface Morphology in High Cycle Fatigue Failure of IN718: Orion Kafka1; Jake Benzing1; Nikolas Hrabe1; Newell Moser1; Donald Godfrey2; Philipp Schumacher2; Chad Beamer3; 1National Institute of Standards and Technology; 2SLM Solutions; 3Quintus Technologies
     Metal additive manufacturing (AM) is known to produce material with relatively rough surface finish in the as-built condition. It is important to understand the impact such surfaces have upon mechanical performance, especially fatigue performance, because many desired applications for AM involve convoluted shapes with included surfaces that are difficult to post-process. Using X-ray computed tomography, this study explores the roles of complex three-dimensional surfaces within the context of fatigue. Specifically, IN718 samples were built via laser powder bed fusion using a range of different processing parameters, thereby driving different surface roughness characteristics and fatigue behavior. Geometric and microstructural features most conducive to failure were isolated using a combination of experimental and computational techniques, with particular emphasis towards failures driven by surface conditions. The goal of the research is to provide guidance to machinists and manufacturers towards maximizing fatigue life in the presences of adverse, as-built conditions.

3:20 PM  
The Influence of Orientation and Processing Method on Fatigue Crack Growth Behavior of AM Stainless Steel: Christine Smudde1; Michael Hill1; Christopher San Marchi2; Jeffery Gibeling1; 1Univeristy of California, Davis; 2Sandia National Laboratories
    Additive manufacturing (AM) offers technological advancements supporting innovative engineering design, but also introduces challenges in fatigue critical applications. Due to highly localized heating and the resulting temperature gradients, AM parts often have significant residual stress and distinctive grain patterns that contribute to unique mechanical behaviour. In this study, fatigue crack growth resistance of AM Type 304L stainless steel produced by both directed energy deposition and powder bed fusion was evaluated for crack growth parallel and perpendicular to the build direction. Decreasing alternating stress intensity factor tests were used to assess fatigue crack growth behaviour in the near threshold regime. Constant applied alternating stress intensity factor tests were used to reveal variations in fatigue crack growth rates attributed to residual stress and microstructural contributions for both materials. Finally, assessments of microstructure through large area SEM images and crack path profiles revealed non-uniform grain morphology and varying levels of tortuosity, respectively.

3:40 PM  
Mesoscale Modeling of the Additively Manufactured 316L: Effects of Microstructure and Microscale Residual Stresses: Mohammadreza Yaghoobi1; Yin Zhang2; Krzysztof S. Stopka2; David J Rowenhorst3; Ting Zhu2; John E. Allison1; David L. McDowell2; 1University of Michigan; 2Georgia Institute of Technology; 3US Naval Research Laboratory
    Microstructure and residual stresses play a key role in the response of components produced by Additive Manufacturing (AM). While macroscopic residual stresses are commonly considered in computational models, the effect of micro-residual stress is less well established. The current work investigates the effects of microstructure and micro-residual stresses on the response of AM stainless steel. A crystal plasticity model is developed as an extension of the Armstrong–Frederick cyclic hardening plasticity model which captures the tensorial evolution of backstress induced due to micro-residual stresses. Open source PRISMS-Plasticity and PRISMS-Fatigue software are used to conduct large scale CPFE simulations and fatigue analysis. Reconstructed AM stainless steel microstructures are used to investigate the interaction of micro-residual stresses with the microstructural features on cyclic responses including stress-strain and elastic lattice strain. The effect of micro-residual stresses and microstructure on the microscopic response including the local plastic slip and fatigue driving forces is addressed.

4:00 PM Break

4:20 PM  Invited
Effect of Stress State and Pores on Multiaxial Fracture of Low- and High-ductility Additively Manufactured Metals: Allison Beese1; 1Pennsylvania State University
    The effect of internal pores and stress state on the fracture behavior of relatively high ductility 316L stainless steel and relatively low ductility Ti-6Al-4V manufactured via laser powder bed fusion (L-PBF) additive manufacturing (AM) was investigated. To do so, penny-shaped pores, with varying diameters, were intentionally fabricated inside cylinders during the layer-by-layer AM process. Uniaxial tension and notched tension specimens were machined from these cylinders, and samples were tested under tensile loading to simultaneously probe the effect of pore size and stress triaxiality (proportional to negative pressure) on the measured ductility of each of the metal alloys. Companion finite element analysis simulations of each test were performed to identify the evolution of stress state and strain accumulation up to fracture. This presentation will describe the fracture models used to capture these data as well as the impact of defects on fracture on these two disparate alloys.

4:50 PM  
Flaw Type Dependent Tensile Properties of 316L Stainless Steel Manufactured by Laser Powder Bed Fusion: Nathalia Diaz Vallejo1; Ke Huang2; Christopher Kain2; Le Zhou3; Jeongmin Woo1; Nicolas Ayers1; Asif Mahmud1; Erin Honse4; Han Chan4; Alexander Hall4; František Zelenka4; Yongho Sohn1; 1University of Central Florida; 2Siemens Energy; 3Marquette University; 4Thermo Fisher Scientific
    Effects of flaws, namely keyhole porosity (KHP) and lack-of-fusion flaws (LFF), on the tensile behavior were examined for 316L stainless steel manufactured by laser powder bed fusion (LPBF) by intentionally introducing them through variations in laser scan speed and laser power, while hatch spacing (0.12mm) and slice thickness (0.03 mm) were held constants. Optical microscopy, electron microscopy and X-ray tomography were used to document amount, type, size, shape and distribution of these flaws. Quasi-static tensile behavior was investigated using both the miniaturized and standard size specimens on these samples with quantified flaw characteristics. Changes in yield strength, tensile strength and elongation at fracture were examined to correlate characteristics of flaws, fracture surface features, relative density of 316L samples and size of the tensile specimens.

5:10 PM  
High Cycle Fatigue Behavior of Additively Manufactured Thin Wall Inconel 718 (Dependence on Thickness and HIP): Paul Paradise; Anushree Saxena1; Andrew Sarrasin1; Nikki Van Handel1; Dhruv Bhate1; 1Arizona State University
    Additively Manufactured Thin-wall Inconel 718 specimens commonly find application in heat exchangers and Thermal Protection Systems (TPS) for space vehicles. The wall thicknesses in applications for these components typically range between 0.03-2.5mm. This study aims at investigating the dependence of High Cycle Fatigue (HCF) behavior on wall thickness and Hot Isostatic Pressing (HIP) for as-built Additively Manufactured Thin Wall Inconel 718 alloys. To address this, High Cycle Fatigue Tests were performed on specimens of seven different thickness (0.3mm, 0.35mm, 0.5mm, 0.75mm, 1mm, 1.5mm and 2mm) using a Fatigue Testing Machine. Only half of the specimen underwent HIP, creating data for both HIP and No-HIP specimens. Upon analyzing the collected data, it was noticed that presence of Porosity in No-HIP specimens make them more sensitive to changes in stress. A clear decrease in fatigue strength with decrease in thickness was observed for all specimens.

5:30 PM  
Non-destructive Determination of Single Crystal Elastic Constants in Additively Manufactured Alloys by Bayesian Inference and Resonant Ultrasound Spectroscopy: Jeffrey Rossin1; Patrick Leser2; Kira Pusch1; Carolina Frey1; Chris Torbet1; Stephen Smith2; Samantha Daly1; Tresa Pollock1; 1University of California Santa Barbara; 2NASA Langley Research Center
    Qualification of additively manufactured components has limited the usage of these components in critical applications. Knowledge of both the single crystal elastic constants and texture of built AM microstructures is necessary for designing the anisotropic (as-built) components, but often unknown for AM materials. Determination of the single crystal elastic constants typically requires fabrication of a single crystal. For AM alloy compositions, single crystals are difficult to produce. Resonant ultrasound spectroscopy (RUS) inversion techniques have proven reliable for determining the texture and aggregate elastic constants of AM components, but required single crystal elastic constants as prior knowledge. In this work, the single crystal elastic constants are determined by informing the RUS inversion with polycrystalline texture information. A parallelizable sequential Monte Carlo (SMC) Python package reduces the computational (Bayesian inference) load by an order of magnitude or more. Full probability distributions are obtained for each parameter, resulting in robust uncertainty estimates.