Tresa M. Pollock, University of California, Santa Barbara
Presentation Title: "Challenges in Acquisition of Statistically Significant Multimodal 3D Data for Property Prediction"

 

The development of high fidelity material property and life prediction models often requires three-dimensional information on the distribution of phases, interfaces, grains or extrinsic defects. Acquisition of this information requires consideration of appropriate representative volume elements that vary in size with material, microstructure and the property of interest. Recent advances in the use of femtosecond lasers for layer-by-layer ablation within the TriBeam microscope platform to gather representative volumes of information will be reviewed. Examples of the utility of multimodal data will be given for several classes of materials. Statistical measures of convergence for microstructural features and their relationship to convergence volumes for properties will be considered. The challenges for integrating experimental voxelized data with models for prediction of mechanical properties will be addressed for nickel-base and titanium-base alloys.

About the Speaker: Tresa M. Pollock is the Alcoa Professor of Materials at the University of California, Santa Barbara. She graduated with a B.S. from Purdue University in 1984, and a Ph.D. from Massachusetts Institute of Technology in 1989. Pollock was employed at General Electric Aircraft Engines from 1989 to 1991, where she conducted research and development on high temperature alloys for aircraft turbine engines. She was a professor in the Department of Materials Science and Engineering at Carnegie Mellon University from 1991 to 1999 and the University of Michigan from 2000 to 2010. Her current research focuses on processing and properties of structural materials and coatings and on the use of ultrafast lasers for microfabrication, tomography and materials diagnostics. Pollock was elected to the U.S. National Academy of Engineering in 2005, is a Fellow of TMS and ASM International, serves as principal of the Metallurgical and Materials Transactions family of journals, and was the 2005 TMS President.

Henry Proudhon, Mines Paristech, Centre des Matériaux
Presentation Title: "Polycrystalline Materials in 4 Dimensions"

Non-destructive synchrotron 3D microstructural imaging is now mature and can be paired with mechanical loading (4D testing) and subsequent mechanical simulations at the grain scale. This is a key to advance our understanding of how the polycrystalline microstructure controls the mechanical properties of structural materials. In the last years, large efforts were devoted to develop 3D orientation mapping capabilities in mm sized specimens, with micrometer spatial resolution. Automated serial sectioning methods paired with EBSD 3D orientation mapping, can target very complex microstructures although at the price of destructing the specimen. In parallel, the increasing popularity and capabilities of hard X-ray tomography coupled to diffraction to image the bulk of materials in three dimensions brings forward a new way to conduct microstructurally informed mechanical testing. In particular, Diffraction Contrast Tomography now provides 3D grain maps non destructively and allow further crystallographic specific investigations during mechanical testing. One of the key challenges is then to link 3D microstructure characterization tools with computational models (eg by finite elements or FFT) to predict engineering properties such as strength or fatigue resistance.

In this presentation, examples of 3D experimental microstructure based large-scale computations using the crystal plasticity finite element method will be presented and compared with in situ mechanical testing experiments. Examples with microstructures obtained with both serial sectioning and Diffraction Contrast Tomography will be targeted. Simultaneous modeling/experimental approaches will be discussed in light of the results. One recurring difficulty in the field of 4D studies is the very small number of tested samples, due to the limit of synchrotron beam time availability and to the inherent difficulty to manipulate 4D data sets. Possible solutions to these problems will be discussed.

About the Speaker: Henry Proudhon graduated in 2001 from École Centrale de Lyon in France in mechanical engineering. He received his Ph. D. in material science in 2005 from INSA Lyon working on investigating fatigue cracking mechanisms with synchrotron x-ray tomography. He then went to the University of British Columbia to work with Warren Poole on ultrafine-grained materials and the Bauschinger effect. In 2007, he joined CNRS, the French national research institute for science, at Centre des Matériaux MINES ParisTech to carry out his research on three dimensional study of deformation and fracture in polycrystalline materials: from synchrotron x-ray investigations to computational mechanics. In 2015 he defended his habilitation thesis and also was associated to the DiffAbs beamline at the SOLEIL synchrotron near Paris. Last year he was a visiting researcher at the University of California, Santa Barbara in Tresa Pollock’s research group, to work on Ni-based superalloys and TiAl alloys. Back in Europe, he continues to work to improve structural materials with partners like Safran, Dassault Aviation or Arkema. He has coauthored more than 50 publications in international peer reviewed journal and received several award such as the FEMS/TMS Young Leaders International Scholar award in 2016.

Stéphane Roux, National Center for Scientific Research
Presentation Title: "Imaging Mechanical Models"

Digital Volume Correlation (DVC) consists in a non-rigid registration of 3D images of materials. When successive images of a specimen under load are processed, space-time displacement can be measured, and from them mechanical properties can be inferred through mechanical identification. Tomographic image reconstruction, DVC and identification are three inverse problems that share a number of similarities. Moreover they can be combined by pairs and even into a single procedure with radiographs as input and mechanical properties as output. The fusion of these operations results in a net gain in fidelity and efficiency, allowing kinematic or mechanical models to be used as regularizations to help solving the inverse problem. Additionally, it also opens the door to model reduction techniques to further optimize the entire process. This “grand unification” will be illustrated through different examples showing that a mechanical test classically performed in one week, can be completed in a few minutes with a similar quality.

About the Speaker: Stéphane Roux graduated from École Polytechnique in 1983 and the École Nationale des Ponts et Chaussées (ENPC) in 1985. He received his Ph.D. in mechanical engineering from the ENPC in 1990. As a CNRS research professor, he was successively at École Supérieure de Physique et Chimie Industrielles de la Ville de Paris (ESPCI), at the joint CNRS/Saint-Gobain Research Laboratory, and currently, he is at the Laboratory of Mechanics and Technology at École Normale Supérieure de Paris-Saclay. Roux holds seven patents, and is the author of more than 350 publications. In 2006, he received the Silver Medal in Information and Engineering Sciences from the CNRS. His current research activities are in theory at the service of experimental mechanics—the development of image-based measurements in solid mechanics, aimed to deliver quantitative evaluation of mechanical properties with the least uncertainties.

5700 Corporate Drive Suite 750

Pittsburgh, PA 15237

Telephone:

U.S. and Canada Only: 1-800-759-4867

Other Countries: 1-724-776-9000, ext. 241

Fax: 1-724-776-3770

E-mail: mtgserv@tms.org