The Laboratory in the News

Record Simulations Conducted on Sequoia

In January 2013, Livermore researchers completed the largest particle-in-cell (PIC) simulations ever performed, using all 1,572,864 cores of the Sequoia supercomputing system. Sequoia is an IBM BlueGene/Q machine and the first supercomputer to exceed 1 million computational cores. At peak operation, it can process 20 quadrillion floating-point operations per second.

Frederico Fiuza, a Lawrence Fellow and physicist in the Laboratory’s Physical and Life Sciences Directorate, led the simulation effort, which used the OSIRIS code to examine the fast-ignition approach to sustained thermonuclear burn and energy gain (simulation shown at right). In fast ignition, lasers deliver more than 1 petawatt (a million billion watts) of power in less than one-billionth of a second to heat compressed deuterium–tritium fuel to temperatures exceeding 50 million degrees Celsius.

Fast ignition differs from the central hot-spot approach used at the National Ignition Facility in which laser beams simultaneously compress and ignite a spherical fuel capsule. Instead, fast ignition adds a high-intensity, ultrashort-pulse laser to “spark” ignition.

OSIRIS has been developed over more than 10 years in a collaboration involving the University of California at Los Angeles and Instituto Superior Técnico in Portugal. In the record-setting simulations, the code demonstrated excellent scaling on the full Sequoia system. OSIRIS operated at 75-percent efficiency under “strong” scaling, in which a relatively small problem of fixed size is modeled with a large number of cores. It performed even better under “weak” scaling, achieving 97-percent efficiency when the total problem size was increased.

Fiuza notes that processing these large problems can take an entire year on a cluster of 4,000 cores, but Sequoia can produce results in one day. “We can also simulate problems 400 times greater in size in the same amount of time,” he says. “Combining this unique supercomputer with the highly efficient and scalable OSIRIS code is allowing for transformative research.”

Breakthrough Research on a Deadly Parasite

An international collaboration involving Livermore physicist Matthias Frank and postdoctoral researcher Mark Hunter has determined the atomic-scale structure of a protein that is key to the survival of the single-celled parasite Trypanosoma brucei. T. brucei is responsible for African sleeping sickness, which kills 30,000 people each year and debilitates many more. The team’s research appeared in the January 11, 2013, issue of Science and was recognized as one of the top 10 science breakthroughs in 2012.

Traditional x-ray diffraction studies could not fully characterize the propeptide, or precursor form, of this essential protein, because the enzyme’s crystals were not large enough for analysis. The collaboration instead used the Linac Coherent Light Source (LCLS) at SLAC National Accelerator Laboratory. In a technique called diffraction before destruction, individual nanometer-size crystals produced by the parasite are passed one by one through the LCLS x-ray beam. The resulting diffraction data on 178,875 nanocrystals were then “stacked” for analysis.

For this project, Frank and Hunter developed the nanoparticle injectors, set up the laser pump probe experiments, prepared samples, modeled damage, and monitored data acquisition during the LCLS experiments. Researchers will use the structural information obtained in the study to develop drugs that mimic the propeptide, inhibiting the enzyme and thus killing the parasite. The team’s achievement also demonstrates that diffraction before destruction is a viable approach for obtaining biomolecular information in LCLS experiments.

The Chemistry of Boron at Room Temperature

Livermore researchers Tadashi Ogitsu and Eric Schwegler, along with Giulia Galli of the University of California at Davis, have for the first time characterized the element boron at room temperature. In the March 8, 2013, online edition of Chemical Reviews, the team describes the history of boron research and details the properties of beta boron—the element’s stable form at room temperature—as inferred from experiments and theories.

Boron remains the only element purified in macroscopic quantities for which the ground-state geometry has not been completely determined by experiments. Theoretical progress over the last decade has revealed numerous properties of elemental boron, allowing researchers to thoroughly characterize its structure at ambient conditions as well as its electronic and thermodynamic properties.

Boron sits on the first row of the periodic table in a peculiar, transitional position. The only nonmetal in the third column of the table, it is flanked by metallic elements on its left and nonmetals on its right. The crystallographic structure and topology of beta boron are extremely complex. Beta boron is characterized by interconnecting icosahedra—a regular polyhedron with 20 identical equilateral triangular faces—and partially occupied sites. It also has more than 300 atoms per unit cell, an unusually large number. No other element on the periodic table shares these features.