Experiments at the Scale of Simulations
TO advance our understanding of materials, especially when they are subjected to extreme conditions, Lawrence Livermore researchers are combining theory, experiments, and simulations in unprecedented ways. Our scientists and engineers are using the newest generation of the National Nuclear Security Administration’s massively parallel supercomputers and codes based on first principles (the interactions between atoms) to generate atomistic simulations of material properties and performance at extremely short time and length scales. The data generated by these simulations allow scientists to develop a fundamentally new understanding of how materials respond to extreme temperatures and pressures.
Until recently, however, experimentalists could not carry out experiments at extreme conditions that simultaneously explored the length and time scales on which atoms move and the physics of the material response are controlled—that is, the scale of atomistic simulations. Typically, insights came from analyzing a material before and after it was subjected to extreme conditions. Scientists then inferred the governing phenomena from observations. Therefore, it was natural to ask: Are there ways in which we can experimentally probe these atomistic length and time scales simultaneously to both improve our understanding and validate our simulations?
As described in the article A New Realm of Materials Science, new techniques are allowing us to expand the amount of experimental information we can obtain on shocked materials at the same time and length scales used in simulations. The first technique, dynamic x-ray diffraction, requires high-intensity lasers to illuminate the changing microstructure of crystalline solids. For these experiments, researchers are using lasers at Lawrence Livermore and Los Alamos national laboratories, the University of Rochester, and Rutherford Appleton Laboratory in the United Kingdom.
The second technique, x-ray scattering, uses high-intensity x-ray sources generated by an accelerator light source, or synchrotron, to reveal defects and voids in microstructures. The Department of Energy’s Office of Science is constructing the world’s brightest x-ray source, called the Linac Coherent Light Source, at the Stanford Linear Accelerator Center. We plan to take advantage of the facility when it is completed in 2009.
By using dynamic x-ray diffraction and scattering techniques, Livermore researchers for the first time will be able to directly compare the results of experiments and simulations. The researchers, funded by the Laboratory Directed Research and Development Program, are drawn from the Defense and Nuclear Technologies, Chemistry and Materials Science, National Ignition Facility (NIF) Programs, Engineering, and Computation directorates. We also have partnered with colleagues at other national laboratories and universities, including University of California campuses.
The new techniques will bring the experimentalist and computer scientist even closer together. Experiments allow us to plan and refine simulations, while simulations guide us to look for certain events in the results. Data gained from both approaches strengthen our theories and models.
This new era of experimental research is extremely important for the nation’s stockpile stewardship program because material behavior is at the heart of most issues associated with this mission. In particular, scientists want to improve their ability to predict the effects of aging on weapon parts or the likely performance of manufactured parts. The benefits of this research are certain to
be much broader. For example, I expect that an important payoff will be for the NIF Programs, which plans to use the world’s most energetic laser, located at Livermore, to explore the properties
of materials driven to extreme conditions.
I also anticipate that we will further our understanding of how materials change phase, how defects in crystalline solids evolve, how a metal’s microstructure changes in time, and how and why various materials fail. The result will yield important contributions to stockpile stewardship, materials science, industry, and basic scientific understanding.