A team of researchers led by Marco Govoni attended the Argonne GPU Hackathon, a multi-day event pairing domain scientists with experienced GPU mentors to help developers accelerate their own codes on graphics processing units (GPUs). Several MICCoM codes are currently being optimized for state of the art GPU architectures.
List of talks:
Marco Govoni, Francois Gygi, and Giulia Galli have been awarded 1,200,000 Theta node-hours to advance the design of novel materials for both solar cell devices and quantum platforms, including sensors. The award is through the U.S. Department of Energy's (DOE) Office of Science Innovative and Novel Computational Impact on Theory and Experiment (INCITE) program, which has awarded allocations of supercomputer access to 51 high-impact computational science projects for 2021. The simulations of electronic excited state properties of heterogeneous materials—including defects and interfaces—couple first-principles molecular dynamics and electronic structure methods beyond density functional theory, as implemented in the Qbox and WEST codes open-source code.
The article "Electron affinity of liquid water", which employed the WEST code and was featured by Argonne news, appeared as part of a themed collection on "Computation and Machine Learning for Chemistry" by the journal Nature Communications. The open-access collection is a showcase of exceptional recent content in the category of computational chemistry.
Marco receives the DOE Early Career award from the U.S. Department of Energy, Office of Science, Basic Energy Sciences, Theoretical Condensed Matter Physics program. The awarded project is titled "Optical Control of Spin-polarization in Quantum Materials", and aims to enable the development of robust solid-state qubits as building blocks for applications in quantum sensing, quantum communication, and quantum computing. This highly competitive award supports the computational study of electronic states of defects in semiconductors. The project harnesses pre-exascale simulations that incorporate elements of machine learning and quantum computing in order to study optically activated processes and understand how the local environment tunes the properties of spin defects. [Argonne news]
The paper entitled "Quantum simulations of materials on near-term quantum computers" has been accepted on npj Computational Materials. We wrapped up calculations performed on a quantum computer and showed that they yield results in agreement with those obtained with exact diagonalization on classical architectures, paving the way to simulations of realistic materials on near-term quantum computers. Our hybrid classical-quantum method is based on a quantum embedding theory for the calculation of strongly-correlated electronic states of active regions, with the rest of the system described within density functional theory. We demonstrate the accuracy and effectiveness of the approach by investigating several defect quantum bits in semiconductors that are of great interest for quantum information technologies. [Argonne news]
Marco is teaching the class Applied Scientific Computing in Molecular Engineering at the University of Chicago. This one-quarter course provides hands-on practical training in scientific computing with a focus on applications to molecular engineering. The course provides: (i) training in core programming concepts, including a broad introduction to Python programming and use of key scientific libraries, (ii) advanced programming topics in CPU and GPU parallel programming and quantum computing, exploring their use through practical examples. Hands-on immersive praxis, mostly using electronic notebooks, will introduce students to the efficient use of several computational resources such as pre-exascale and quantum computers.
The project, "Advanced Materials Characterization with AI-informed Computation," led by Marco has been selected for the 2019-2020 Argonne Leadership Computing Facility Data Science Program (ADSP) at Argonne National Laboratory. ADSP connects leading researchers with ALCF scientists to push the state-of-the-art in machine learning, workflows, data analysis, and algorithmic development. The awarded project will aim to develop an AI-informed computational framework that integrates machine learning into existing workflows, thereby enabling the prediction of spectroscopic signatures of materials with fast turnaround.
The U.S. Department of Energy announced that it will invest $32 million over the next four years to accelerate the design of new materials through use of supercomputers. Marco is co-Principal Investigator of MICCoM, one of the seven awarded centers.
The University of Chicago announced a $100 million commitment from the Pritzker Foundation to support the new Pritzker School of Molecular Engineering, becoming the first university in the United States with a school dedicated to this emerging field. The new school builds on the success of efforts by the University and Argonne National Laboratory, which partnered in 2011 to establish the Institute for Molecular Engineering. Marco will work closely with the new school on research collaborations.
Marco presented the highlight talk entitled "Qresp, a tool for curating, discovering and exploring reproducible scientific papers" at EPFL. The keynote focused on outstanding problems (and proposed solutions) in sharing reproducible workflows in computational materials science. Qresp has previously been featured by UChicago News.
Marco joins the Chicago Quantum Exchange, a growing intellectual hub for the research and development of quantum technology in the Chicago area.
Marco contributed to the symposium Sustainable Software for Computational Molecular Science, organized by MolSSI at the Spring 2019 ACS Meeting in Orlando from March 31-April 4, 2019. This program brought together leading researchers and software developers from academia and industry in the computational molecular sciences for a 5 day, 9 session meeting.
The project lead by M. Govoni has been accepted as a Tier 1 NESAP application. The National Energy Research Scientific Computing (NERSC) Center has announced the latest round of NERSC Exascale Science Application Program (NESAP) teams that will focus on simulation, data analysis, and machine learning applications to prepare workloads for NERSC’s next supercomputer, Perlmutter.
List of talks:
Thanks to the Hour of Code initiative, I met incredibly smart and motivated kids! Who said that computer science and quantum physics cannot be explained to elementary school students? Imagination is more important than knowledge! (A. Einstein).
Leading experts in quantum research from industry, government and universities will gather at the University of Chicago on Nov. 8 and 9 for a summit on this emerging technology and the global race to develop quantum’s revolutionary potential. The Chicago Quantum Summit, which will include technology leaders at Microsoft, IBM, Alphabet Inc.’s Google, the National Science Foundation, the U.S. Department of Energy, the U.S. Department of Defense and the U.S. National Institute of Standards and Technology, will be hosted by the Chicago Quantum Exchange.
UChicago's Library is working with us to explore ways in which Qresp can potentially integrate with the Library’s new institutional repository platform. [news].
Our paper, The role of defects and excess surface charges at finite temperature for optimizing oxide photoabsorbers, is published in Nature Materials! Three major factors determine the chemical reactivity of WO3 interfaced with water: the presence of surface defects, the dynamics of excess charge at the surface, and finite temperature fluctuations of the surface electronic orbitals.
Marco is part of the Quantum Information Science and Engineering Network (QISE-NET), which is housed at the Chicago Quantum Exchange. QISE-NET is an intellectual hub and partnership for advancing academic and industrial efforts in the science and engineering of quantum information. Marco is hosting at Argonne National Lab Tyler Smart, graduate student (University of California, Santa Cruz), with the purpose of developing numeric techniques to model quantum materials for quantum information science.
Argonne National Lab is funding Marco's LDRD research focused on developing a programmable cyberinfrastructure for distributed electronic structure calculations that will fill the gap between researchers and leadership computing facilities.
Invited talk on "Coupling first principles molecular with advanced sampling and many body perturbation theory codes" at the Symposium on Materials Genome Towards Exascale, held in Spetses (Greece), June 10-15 2018.
Molecular engineering scales up to lab-wide research enterprise. The full story appeared in Argonne Today.
Invited talk on a wide range of method developments in electronic structure! Opening the session: Petascale Science and Beyond: Applications and Opportunities for Materials, Chemical, and Bio-Physics III.
Contributed talk on reproducibility in Science! The open-source suite Qresp is expected to increase experimental and computational rigor in reporting results, together with transparency and to greatly facilitate paper reproducibility.
List of talks: