RCAC cluster Negishi powers study of additively manufactured materials
Researchers at Purdue are studying the behavior of additively manufactured (otherwise known as 3D-printed) materials, thanks to the powerful capabilities offered by the Rosen Center for Advanced Computing (RCAC)’s Negishi cluster.
Krzysztof Szymon Stopka, a research engineer and postdoctoral research associate working under Michael Sangid, professor of aeronautics and astronautics and director of the Advanced Computational Materials and Experimental Evaluation (ACME) Lab, is involved in a project focused on advancing the qualification of additively manufactured materials, particularly for structural applications in the aerospace field.
Sangid’s research group combines knowledge of materials science, solid mechanics, and advanced manufacturing to solve complex problems in materials behavior and processing. They employ physics-based computational modeling and design tools, which are experimentally validated and verified. The goal of this work is to improve our understanding and our tools for designing, processing, and “lifing” (testing the durability and expected useful life of) materials through simulation-based modeling of the material structure at the microscale, known as microstructure.
Negishi is helping Stopka’s team mimic and understand the performance of additively manufactured materials. The researchers use Negishi to perform computational modeling to supplement their experimental test campaigns, which are indispensable but costly and time-consuming [1]. Negishi's remarkable speed and reliability make it the preferred choice for these simulations, ultimately enhancing the cost-effectiveness and efficiency of the research [2].
"Negishi is a tremendous resource to our research group,” Stopka enthuses. Negishi is his preferred choice for both batch submissions and interactive sessions. Stopka's project, sponsored by the National Institute of Standards and Technology (NIST), aims to increase the rate at which additively manufactured materials are qualified and implemented in industrial applications [3]. "The computational demands of our research are immense," Stopka states. "Some simulations require nearly 200 CPU cores and can take weeks to complete. Given the volume of simulations needed for our work, RCAC’s clusters are indispensable."
Negishi, which is optimized for traditional, tightly-coupled science and engineering applications, was built through a partnership with Dell and AMD in 2022. Negishi consists of 460 Dell PowerEdge compute nodes with two 64-core 3rd Generation AMD EPYC “Milan” processors (128 cores per node) and 256 GB of memory, six large memory nodes with 1 TB of memory, and 15 AMD Instinct M1210 GPUs.
Negishi is named in honor of the late Ei-ichi Negishi, the Herbert C. Brown distinguished professor of chemistry and the winner of the 2010 Nobel Prize in chemistry. The Negishi supercomputer was dedicated in a February 2023 ceremony featuring campus leaders including Purdue President Mung Chiang and members of the Negishi family.
Many of the graduate student researchers working with Sangid and Stopka in the ACME lab actively use the RCAC clusters for their research. Brandon Mackey is unraveling the complexities of thermo-mechanical fatigue in aerospace alloys. Leonidas Zisis is examining hydrogen embrittlement in Nickel. Kyle Jung, Harshit Gaddam, and Luca Loiodice are addressing other critical problems with additive manufactured materials. Javi Solano and Marco Zambolin use the RCAC clusters to analyze data-rich experimental results. Sean Skweres is simulating high temperature carbon fiber composites to uncover damage mechanisms.
To learn more about Negishi and other RCAC resources, contact rcac-help@purdue.edu.
References:
[1] K. S. Stopka, A. Desrosiers, T. Nicodemus, N. Krutz, et al., "Intentionally seeding pores in additively manufactured alloy 718: Process parameters, microstructure, defects, and fatigue," Addit. Manuf., 66, 103450 (2023)
[2] K. S. Stopka and M. D. Sangid, "Modeling fatigue behavior of additively manufactured alloys with an emphasis on pore defect morphology," J. Mech. Phys. Solids, 181, 105429 (2023)
[3] K. S. Stopka, A. Desrosiers, A. Andreaco, M. D. Sangid, "A methodology for the rapid qualification of additively manufactured materials based on pore defect structures," Integr. Mater. Manuf. Innov., (accepted, in press)