Three Argonne researchers will discuss their Laboratory-Directed Research and Development (LDRD) sponsored work at the LDRD Seminar Series presentation Tuesday, May 14, 2019, at 12:30 p.m. in Building 212, Room A157. All are welcome to attend.
Visit the LDRD website to view upcoming seminars.
Tao Li
“Formation of Polymer-Protein Core-Shell Structures through Controlled Assembly” by Tao Li (XSD)
Abstract
In this talk, we develop a method via controlled assembly of polymers and proteins into a hierarchical structure, which can preserve protein conformation and functionality. Compared to other methods, this proposed work is more straightforward and has better control of the structural features as well as the physical properties. It is important to note that the methods developed in this proposal for the guided assembly of proteins will be broadly applicable for many enzymatic systems.
Biography
Tao Li got his Ph.D. from University of South Carolina in 2009. He then joined Argonne as postdoc in 2010 and promoted to assistant scientist in 2013. He is an assistant professor in the Department of Chemistry and Biochemistry at Northern Illinois University and holds a joint position at Argonne. His research interests are materials, chemistry, energy, biomedicine and synchrotron radiation. He has published more than 70 papers in peer review journals and has been invited to give more than 40 representations at international conferences. Li is also a guest editor of the journal Frontiers in Chemistry.
John Power
“Emittance Exchange for High Brightness Beams,” by John Power (HEP)
Abstract
Methods to control the electron bunch transverse phase space have long been available to the accelerator designer yet methods for controlling the longitudinal phase space are scarce, but essential, for a variety of future accelerator applications. In particular, direct control over the longitudinal position coordinate is extremely difficult due to the exceedingly short duration (femtosecond to picosecond range) of a typical electron bunch. Existing methods for manipulating the longitudinal properties of the bunch (energy spread, bunch length, energy chirp, linearity and emittance) allow for relatively limited control. This LDRD project focuses on developing a new method at Argonne for manipulating the longitudinal properties of a bunch that have significantly improved capabilities over existing methods.
Biography
John Power is an accelerator beam scientist working at the Argonne Wakefield Accelerator facility in the High Energy Physics division. His research interests include a variety of wakefield accelerator physics topics including collinear wakefield acceleration, dielectric two-beam acceleration, enhanced transformer ratio wakefield acceleration, multimode wakefield acceleration. Other areas include high-power testing and modeling of dielectric loaded accelerating structures (multipactor, joint breakdown), the physics of electron beam generation and propagation especially RF photocathode gun and linac design, simulation and fabrication.
Paul Romano
“Rethinking Parallelism in Monte Carlo Neutron Transport Simulations,” by Paul Romano (MCS)
The Monte Carlo method is widely used to simulate neutron transport in nuclear reactor systems due to its accuracy and lack of approximations but requires a longer time-to-solution than other more approximate methods. While the MC method can easily take advantage of multi-core/multi-node systems to reduce time-to-solution, the algorithm normally used to distribute work among multiple processes or threads generally results in poor use of the computing resources, namely the memory hierarchy. While other algorithms have been proposed to make better use of resources, most previous studies have been aimed at taking advantage of data-level parallelism rather than better use of the memory hierarchy. In this talk, we report on a study of the potential for alternate parallel algorithms to improve the temporal and spatial locality of memory accesses in Monte Carlo neutron transport simulations.
Biography
Paul Romano is a computational scientist in the Mathematics and Computer Science (MCS) division with expertise in nuclear reactor physics, nuclear data, high-performance computing and software development. He completed his Ph.D. in nuclear science and engineering at the Massachusetts Institute of Technology in 2013 and later joined Argonne as a staff member in 2015. Romano is the original author and maintainer of OpenMC, an open-source framework for Monte Carlo particle transport that he started during his graduate studies. OpenMC has become a central component in MIT’s research in reactor physics and is used by various universities and laboratories around the world. OpenMC has also served as the primary platform for advanced R&D on Monte Carlo methods within MCS and is starting to be used for production calculations by researchers in the Nuclear Science and Engineering division. Romano is also a lecturer in the masters of computer science program at the University of Chicago.
His research focuses on the performance, accuracy and usability of particle transport simulations, particularly for large-scale nuclear reactor calculations. As part of the Exascale Computing Project, he is leading efforts at Argonne to enable high-fidelity multiphysics simulations of small modular nuclear reactors by coupling the OpenMC code to Nek5000, a computational fluid dynamics code.