Additive Manufacturing: Advanced Characterization with Synchrotron, Neutron, and In Situ Laboratory-scale Techniques: Structure and Microstructure
Sponsored by: TMS: Additive Manufacturing Committee
Program Organizers: Fan Zhang, National Institute of Standards and Technology; Tom Stockman, Los Alamos National Laboratory; Tao Sun, Northwestern University; Donald Brown, Los Alamos National Laboratory; Yan Gao, Ge Research; Amit Pandey, Lockheed Martin Space; Joy Gockel, Wright State University; Tim Horn, North Carolina State University; Sneha Prabha Narra, Carnegie Mellon University; Judy Schneider, University of Alabama at Huntsville

Wednesday 2:00 PM
February 26, 2020
Room: 8
Location: San Diego Convention Ctr

Session Chair: Fan Zhang, National Institute of Standards and Technology


2:00 PM  Invited
Non-destructive Quality Evaluation of Additively Manufactured Metal Components: Sam Yang1; Leon Prentice1; Tony Murphy1; Sherry Mayo1; Clement Chu1; Anna Paradowska2; 1CSIRO; 2ANSTO
    Additive manufacturing (AM) is particularly important for high-value applications such as aerospace. However, due to the specific conditions of the AM process, AM components often have internal microscopic structural defects such as porosity that are difficult to detect non-destructively with current off-the-shelf technology. Conventional X-ray CT (computed tomography) imaging analysis is inadequate in resolving the microscopic porosity defects in macro-sized parts. CSIRO has developed a data-constrained modelling (DCM) technology. Together with the quantitative X-ray CT capability, it allows the resolution of fine defects in AM metal components smaller than the imaging voxels (voxel partial porosity). The technology is being further developed for macro-sized AM metal components using industrial X-ray CT facilities. The ANSTO instrument Kowari allows us to non-destructively determine residual stresses and texture within the interior of bulk engineering components. Imaging beamline Dingo and IMBL are used to assess defects and dimensional tolerance of internal features of such components.

2:25 PM  Invited
Unsupervised Learning of Dislocation Motion: Darren Pagan1; Thien Phan2; Jordan Weaver2; Austin Benson3; Armand Beaudoin1; 1Cornell High Energy Synchrotron Source; 2National Institute of Standards and Technology; 3Cornell University
    The application of machine learning to materials characterization data from engineering alloys has been primarily limited to classification of microstructural features and correlation to observed properties. Instead we propose the application of the unsupervised learning technique, locally linear embedding (LLE), to analyze in-situ diffraction data and find lower-dimensional embeddings that characterize microstructural transients. We apply the approach to diffraction data gathered during uniaxial deformation of additively manufactured Inconel 625. With the aid of a physics-based material model, we find that the lower-dimensional coordinates determined using LLE appear to reflect the evolution of the defect densities that dictate strength and plastic flow behavior. The implications of the findings for future constitutive model development and wider applicability to the study of in-situ materials processing, including additive manufacturing will be presented.

2:50 PM  
In Operando X-ray Diffraction during Laser 3D Printing: Samy Hocine1; Helena Van Swygenhoven1; Steven Van Petegem1; Cynthia Sin Ting Chang1; Tuerdi Maimaitiyili1; Gemma Tinti1; Dario Ferreira Sanchez1; Daniel Grolimund1; Nicola Casati1; 1Paul Scherrer Institut
    Selective Laser Melting (SLM) is a well-known process category in Additive Manufacturing in which thermal energy selectively fuses regions of a powder bed. To investigate process parameters for metallic materials in-situ with synchroton X-rays, a miniaturized SLM device has been developped at PSI. The design of the miniaturized SLM device is determined by the requirements for X-ray access and implementability at different beamlines of synchrotron light sources. First in situ X-ray diffraction tests were performed successfully on Ti-6Al-4V samples at the MicroXAS and MS beamlines, located at the Swiss Light Source. By varying the laser power, scanning speed and hatch distance, various energy densities are obtained. The dynamics of the alpha and beta phases during fast heating and solidification are tracked with a time resolution of 50ms. This allows investigating the heating and cooling rates, and the size and shape of the heat affected zone.

3:10 PM  
In-situ TEM Heating Experiments to Study the Effect of Thermal Gradients on Additively Manufactured Ti-6Al-4V Builds: Sriram Vijayan1; Meiyue Shao1; Chris Blackwell1; Sabina Kumar2; Sudarsanam Babu2; Joerg Jinschek1; 1The Ohio State University; 2University of Tennessee, Knoxville
    In the electron-beam melting powder-bed fusion (EBM-PBF) additive manufacturing (AM) process, the uppermost build layer undergoes rapid solidification and reduced partitioning. Additionally, the subsequent layers below the upper-most build layer are exposed to fast thermal cycling and large thermal gradients across the build plane and build direction, resulting in a non-equilibrium microstructure. The sequence of solid-state transformations that leads to the formation of this microstructure is still unclear. In this work, we use a membrane-based in situ TEM heating holder to study solid-state transformations in AM Ti-6Al-4V samples. These TEM heating holders allow rapid heating/cooling of samples in a controlled manner, capturing dynamic processes in real-time at high spatial resolution, thereby simulating various thermal phenomena in the EBM-PBF process. Here, we present results from in situ TEM heating experiments performed on Ti-6Al-4V powder particles and AM parts to understand the sequence of transformations observed during the EBM-PBF process.

3:30 PM  
Observing the Phase Evolution During Selective Laser Melting of a High-Fe β-Ti Alloy from Elemental Powders via In-Situ Synchrotron X-Ray Diffraction: Farheen Ahmed1; Samuel Clark2; Chu Lun Alex Leung2; Yunhui Chen2; Lorna Sinclair2; Sebastian Marussi2; Veijo Honkimaki3; Noel Haynes4; Peter Lee2; Hatem Zurob1; André Phillion1; 1McMaster University; 2University College London; 3European Synchrotron Radiation Facility; 4Collins Aerospace
    Fe is a low-cost alloying element for β-Ti alloys, which have high tensile and fatigue strengths. However, the low cooling rates during casting cause Fe-rich precipitates known as β-flecks to develop during solidification. Using Selective Laser Melting (SLM), one can produce high-Fe β-Ti alloys free of β-flecks as the rapid solidification rates constrain Fe segregation. To design an optimal production route of high-Fe β-Ti alloys, the phase transformation sequence during printing must first be understood. In this study, real-time synchrotron X-Ray Diffraction was employed to characterize phase transformations during the SLM of a high-Fe β-Ti alloy. Infrared images were collected concurrently and converted to temperature. Temperature profiles were matched with the identified XRD peaks to determine the phase evolution. To further reduce costs, elemental powders rather than a pre-alloyed powder were utilized as the starting material.

3:50 PM Break

4:10 PM  Invited
Combining Atom-probe Tomography and Synchrotron Methods to Investigate In-situ Precipitation in AM-produced Alloys: Eric Jaegle1; Philipp Kürnsteiner1; Pere Barriobero-Vila2; Markus Wilms3; Frederic De Geuser4; Dierk Raabe1; 1Max-Planck-Institut Fuer Eisenforschung; 2German Aerospace Center DLR; 3Fraunhofer-Institute for Laser Technology; 4SIMAP - University Grenoble Alpes
    The re-heating induced by adding material track-by-track and layer-by-layer during AM, often termed intrinsic heat treatment, can be used to trigger precipitation reactions in-situ. Due to the inhomogeneous heat input, the state of precipitates (number density, size, chemistry) often varies within the AM-produced specimens. Changing the material composition during deposition for alloy design purposes additionally introduces chemical inhomogeneity. While high-resolution microscopy techniques such as Atom-Probe Tomography (APT) offer detailed insights into e.g. local precipitate chemistry, obtaining a statistically reliable overview of an inhomogeneous material is a laborious process. We therefore combine APT with synchrotron-based methods such as HEXRD and SAXS to achieve a detailed and reliable insight into in-situ precipitation reactions. This information allows to select optimum alloy compositions and to construct kinetic models for the precipitation transformation. The material systems investigated in this study include Al-Sc-alloys and maraging steels (Fe-Ni-Ti/Al), both produced by DED.

4:35 PM  
In-situ TEM Thermal Cycling of AM Steel: Manas Upadhyay1; Eva Héripré2; Lluís Cardona2; Alexandre Tanguy1; Simon Hallais1; Sylvain Durbecq1; Thien-Nga Lê1; 1École Polytechnique; 2CentraleSupélec
    In recent years, increasing research efforts have been dedicated to better understand the origin of far-from-equilibrium microstructure during additive manufacturing (AM) of metals/alloys. Most experiment/modeling efforts aim at studying the role of melt-pool dynamics and rapid solidification on microstructure formation. However, we are interested in studying the solid-state microstructural evolutions occurring after solidification, i.e. during thermal cycling due to the continuation of the building process; large thermo-mechanical driving forces are generated during thermal cycling which result in important microstructural changes. In this talk, we will present some interesting results from a series of in-situ thermal cycling tests performed on AM steel lamellae in a transmission electron microscope. The motivation behind these tests comes from the intractability of following in-situ microstructural changes during AM within the processing chamber. Instead, we reproduce the same thermal cycling, that the material would experience during AM, inside a transmission electron microscope to facilitate in-situ observations.

4:55 PM  
Detection of Early Crack Formation of Fatigued, Additively Manufactured Stainless Steel Using Neutron Dark-field Imaging: Adam Brooks1; Daniel Hussey2; Hong Yao3; Ali Haghshenas3; Jumao Yuan3; Jacob LaManna2; David Jacobson2; Caroline Lowery3; Shengmin Guo3; Michael Khonsari3; Leslie Butler3; 1EWI; 2National Institute of Standards and Technology; 3Louisiana State University
     Abstract: Fatigue in selective laser melted (SLM) and conventionally manufactured stainless steel (SS) 316 dogbones was studied with neutron dark field imaging [1]. To produce the dark field images, a far field interferometer was employed. The dark-field image combines sensitivity to micrometer-sized scattering centers (diameters ranging from ~600 nm to 2000 nm) at crack formation with sub-millimeter image spatial resolution. The crack formation observed with the neutron dark-field was validated post-imaging with additional fatigue cycles to fracture. Further inspection was performed by scanning electron microscopy (SEM) and optical photography. In the two fatigued dogbones, SLM and conventional crack formation was identified to within 1 mm. [1] A.J. Brooks et al, Materials and Design 140 (2018) 420–430.

5:15 PM  Cancelled
Machine Learning Applications for In-situ Synchrotron X-ray Diffraction Measurements of Thermo-Mechanical Behaviors of Additively Manufactured 17-4 Stainless Steel: Thien Phan1; Darren Pagan2; 1National Institute of Standards and Technology; 2Cornell High Energy Synchrotron Source
    Additively manufactured (AM) alloys often exhibit significantly different thermo-mechanical responses than their wrought counterparts. As such, there is a need for new analysis techniques to quantify these responses to accelerate the adoption of these alloys. In this study, we apply unsupervised machine learning, locally linear embedding (LLE), to in-situ X-ray diffraction data to quantify phase transformations and microstructure changes during heating and uniaxial deformation of AM 17-4 stainless steel samples. We correlate the lower-dimensional representation of the diffraction data (from LLE) to material evolution (temperature and phase fraction). Implications for future applications of machine learning for in-situ diffraction experiments during the additive process will be discussed.