Education & Outreach
The Center for Advanced Research Computing collaborates with researchers on their work in numerous ways. In addition to providing computing and storage resources, education, and user support, the CARC also works closely with researchers on their applications for research grants and awards and on their academic publications.
If you're interested in collaborating with the CARC on a grant proposal or crediting the CARC in your publication, please see our Grant and Publication Support page.
The following are recent projects, publications, and awards that the CARC has contributed resources to or collaborated with researchers on.
Grant from National Science Foundation's Campus Cyberinfrastructure (CC*) Program
Title: CC* Compute: A Customizable, Reproducible, and Secure Cloud Infrastructure as a Service for Scientific Research in Southern California
Collaborators: Carl Kesselman (Principal Investigator) and Byoung-Do Kim (Co-Principal Investigator)
In August 2020, the National Science Foundation officially awarded Carl Kesselman, Director of Informatics Systems Research at USC's Information Sciences Institute, and BD Kim, Director of the Center for Advanced Research Computing, with a $400,000 grant for their proposal "CC* Compute: A Customizable, Reproducible, and Secure Cloud Infrastructure as a Service for Scientific Research in Southern California," as part of the Campus Cyberinfrastructure (CC*) Program.
"The Campus Cyberinfrastructure (CC*) program invests in coordinated campus-level networking and cyberinfrastructure improvements, innovation, integration, and engineering for science applications and distributed research projects. Learning and workforce development (LWD) in cyberinfrastructure is explicitly addressed in the program. Science-driven requirements are the primary motivation for any proposed activity."
More information on the grant: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2019220&HistoricalAwards=false
More information on the Campus Cyberinfrastructure (CC*) Program: https://www.nsf.gov/pubs/2020/nsf20507/nsf20507.htm
New cryo-EM facility and collaborations with Dornsife and ISI
Collaborators: Dornsife College of Letters, Arts and Sciences, Amgen, Center for Advanced Research Computing, USC Information Sciences Institute (ISI)
USC Dornsife and Amgen have agreed to house two cryo-EM microscopes at USC, to be operated by the Center of Excellence in Nano Imaging. The CARC is leading the collaborative effort with other ITS teams to create a full research ecosystem for our USC research community, including user interface, data and computational workflow management platform, and special GPU cluster deployment.
The microscopes will transmit up to 16 TB of refined data (images near atomic resolution) per day to USC servers and to Amgen's AWS cloud storage. USC researchers at Dornsife, Viterbi, Keck School of Medicine, and other USC research institutes will make use of the new microscropes for their research.
Publication acknowledgement in American Chemical Society's ACS Catalysis
Title: Pd-Catalyzed Synthesis of Densely Functionalized Cyclopropyl Vinyl Sulfides Reveals the Origin of High Selectivity in a Fundamental Alkyne Insertion Step
Collaborators: Liliya T. Sahharova, Evgeniy G. Gordeev, Dmitry B. Eremin (USC), and Valentine P. Ananikov
On August 19, 2020, the American Chemical Society's ACS Catalysis published "Pd-Catalyzed Synthesis of Densely Functionalized Cyclopropyl Vinyl Sulfides Reveals the Origin of High Selectivity in a Fundamental Alkyne Insertion Step," which included an acknowledgement for work done on the CARC's Discovery cluster by Dmitry B. Eremin.
Publication abstract: We show that the insertion of alkyne into the Pd−S bond proceeds by an asynchronous mechanism, which starts with metal−carbon binding and resolves into diverse transient structures. We further demonstrate that dynamic involvement of these structures ensures regioselectivity of the entire process, thus providing a mechanistic link that has long been missing. Alkyne insertion into the metal−heteroatom bond is a fundamental elementary step and a corner stone of catalysis and organometallic chemistry that works for a large variety of metals and heteroatoms. Mastering its Markovnikov vs anti-Markovnikov selectivity provides powerful opportunities for the design of selective functionalization routes."
Full article: https://pubs.acs.org/doi/10.1021/acscatal.0c02053
Exascale Computing Project - Aurora Early Science Program
Title: Metascalable Layered Materials Genome
Collaborators: University of Southern California (Aiichiro Nakano, Ken-ichi Nomura, and BD Kim), USC CARC (Marco Olguin), Argonne National Laboratory, Intel, and Kumamoto University (Japan)
Project description: A project investigating layered materials (LMs) that hold the promise of producing electricity when exposed to light has been chosen for the Early Science Program on Aurora, the next major exascale supercomputer to launch at Argonne National Laboratory. This project will advance the layered materials genome (LMG). We will perform 105-atom non-adiabatic quantum molecular dynamics (NAQMD) and 1010-atom reactive molecular dynamics (RMD) simulations on Aurora for computational synthesis and characterization of LMs. The large-scale simulations will (1) guide the synthesis of stacked LMs by chemical vapor deposition (CVD), exfoliation and intercalation, and (2) discover function-property-structure relationships in LMs with a special focus on far-from-equilibrium electronic processes.
Tasks: (1) Provide HPC support and consulting; (2) With Intel compiler support for GPU offload now available, implement OpenMP offloading to GPUs for one of the most computationally expensive QXMD kernels in the Local Field Dynamics code.
Grant from National Science Foundation's Future Manufacturing Program
Title: Artificial Intelligence Driven Cybermanufacturing of Quantum Material Architectures
Collaborators: University of Southern California (Aiichiro Nakano and Ken-ichi Nomura), USC CARC (Marco Olguin), Harvard University, and Howard University
Project description: Quantum material architectures consist of graphene and other two-dimensional materials, which, when stacked in precise three-dimensional architectures, exhibit unique and tunable mechanical, electrical, optical, and magnetic properties. These three-dimensional architectures have broad potential applications and are highly promising components for microchips, batteries, antennas, chemical and biological sensors, solar-cells and neural interfaces. This grant is to develop a transformative Future Manufacturing platform for quantum material architectures using a cybermanufacturing approach, which combines artificial intelligence, robotics, multiscale modeling, and predictive simulation for the automated and parallel assembly of multiple two-dimensional materials into complex three-dimensional structures. A key outcome is an AI-driven, robotics-controlled cybermanufacturing microfluidic platform that is capable of manufacturing complex structures for emerging quantum and other device applications.
Tasks: (1) Provide HPC support and consulting; (2) Perform non-adiabatic quantum molecular dynamics simulations on complex, layered three-dimensional structures to determine stacking processes conducive to optimal performance for various device applications.
More information on the grant: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2036359&HistoricalAwards=false
Fluoride Ion Batteries
Title: Synthetic Control Across Length-scales for Advancing Rechargeables (SCALAR)
Collaborators: US Department of Energy (DOE) Energy Frontier Research Center (Class: 2018 - 2022)
USC Collaborators: Brent Melot, Chemistry Department, USC CARC (Marco Olguin)
Project description: The main goal of the project is to better understand the chemical and structural factors that allow for reversible (de)insertion of fluoride ions into intercalation host structures, to develop some rules for how to think about Fluoride Ion Batteries (FIBs) compared to Lithium Ion Battery (LIB) materials. Past work on fluoride transport has studied conversion type electrodes (e.g. converting lead metal into lead fluoride), which have interesting potential, but serious limitations in terms of cycling. Recent advances that might allow for success with reversible FIBs are the demonstration of room-temperature cycling with liquid electrolytes, and then recent work headed by Melot's group that shows F- insertion into ReO3. The problem is that F- is not reversible with ReO3, so learning from this we need to address the side reactions which occur that make it non-reversible, and there is a need to move to layered materials so that the intercalation channels are stable with respect to fluoride de-insertion.
Tasks: (1) Perform ab initio molecular dynamics simulations and reaction barrier calculations based on Density Functional Theory (DFT) using the CP2K and VASP codes to study the complex (de)fluorination mechanisms in FIBs.
More information: http://www.chem.ucla.edu/SCALAR/
Safe electrolytes compatible with both Li-metal anodes and high-voltage cathodes
Title: Developing Safer, Wide-Temperature Liquefied Gas Electrolytes for Lithium Ion Batteries
Collaborators: University of Southern California, USC CARC (Marco Olguin), US Army Research Laboratory, University of California San Diego, and South 8 Technologies
Principal Collaborator: Oleg Borodin, US Army Research Laboratory
Project description: We endeavor to mitigate safety and practicality concerns of high-voltage Lithium Ion Battery electrolytes by demonstrating an enhanced safety feature inherent in liquefied gas electrolytes and by showing the viability of using difluoromethane as a liquefied gas solvent which has lower pressure, lower flammability, and improved maximum operation temperature characteristics compared with fluoromethane. We create a custom-built setup to enable liquefied gas electrolyte characterization through Raman spectroscopy and supplement this with molecular dynamics (MD) simulations. The demonstrated use of such alternative liquefied gas solvents opens a path towards the further development of high-energy and safe batteries that can operate in a wide-temperature range.
Tasks: (1) Perform ab initio molecular dynamics simulations and anode-electrolyte/cathode-electrolyte interface calculations based on Density Functional Theory (DFT) using the CP2K and VASP codes to study the complex interfacial reaction mechanisms of novel liquefied gas electrolytes with lithium metal anodes and high-voltage cathodes.