The evolution and configuration of the matter of which we and our world are formed were started by the Big Bang. Approximately 14 billion years after the Big Bang, Nuclear Physicists from the Oak Ridge National Laboratory (ORNL) of the Department of Energy and their collaborators have used the exceptionally powerful supercomputers in the United States to define the conduct of objects ranging from subatomic neutrons to neutron stars, which drastically differ in terms of size but are closely related to one another by physics.
By means of the DOE Office of Science’s Scientific Discovery through Advanced Computing (SciDAC) program that simultaneously promotes science and supercomputing to quicken discovery, ORNL is taking part in two 5-year computational nuclear physics projects.
Associates in the first project—the Nuclear Computational Low Energy Initiative (NUCLEI)—will compute characteristics and reactions of disparate atomic nuclei, significant in terrestrial experiments as well as astrophysical environments. Nearly 30 Researchers at 12 national labs and Universities are expected to share a funding of $10 million.
Joseph Carlson from Los Alamos National Laboratory (LANL) is the lead of NUCLEI. Stefan Wild from Argonne National Laboratory is the Co-director for Applied Math and Computer Science, and Thomas Papenbrock from the University of Tennessee, Knoxville (UTK) and ORNL is the Co-director for Physics.
A total of 32 Researchers from 12 national labs and universities are part of the second project titled Towards Exascale Astrophysics of Mergers and Supernovae (TEAMS). Using a proposed support of $7.25 million, Researchers will simulate supernovae explosions and neutron-star mergers through which atomic elements weighing more than iron are formed and will also estimate signatures of such cataclysms as gravitational waves. Raph Hix from ORNL has headed TEAMS, in which Bronson Messer from ORNL is the computational lead and Chris Fryer from LANL is the science lead.
“There is a nice synergy—NUCLEI is doing pure nuclear physics and TEAMS is, in a sense, doing applied nuclear physics,” stated Hix, a nuclear astrophysicist. “We need their nuclear physics to do our astrophysics.”
NUCLEI collaborators will compute the structure, interactions, reactions and disintegration of stable and radioactive nuclei—elements that disintegrate into more stable states—for correlating with outcomes of experiments carried out at DOE facilities such as the Facility for Rare Isotope Beams (FRIB) being built at Michigan State University. Since Astrophysicists require high-quality information on the actual behavior of nuclei, information obtained from NUCLEI and from other experiments will be applied in TEAMS simulations that investigate the way in which nuclei are formed under the drastic conditions of dying stars.
In the case of both the projects, Computing and Science Professionals will begin from highly advance models, numerical methods and leadership-class high-performance computers such as Titan, ORNL’s current workhorse supercomputer, or Summit, to be commenced in the year 2018.
Calculating key nuclei
How are neutrons and protons bound by strong force into nuclei? How are neutrons captured by light atomic nuclei to form heavier elements in stars? What are the characteristics of the neutrino, which has a significant role in supernovae explosions and radioactive decay?
Given above are specific questions that the NUCLEI team will try to answer by adopting state-of-the-art Applied Mathematics, Computer Science and Physics to characterize atomic nuclei. The computations are quite expensive.
With 100 or more particles, exact solutions became exponentially costly. New methods enable efficient performance on the fastest supercomputers.
Thomas Papenbrock, the University of Tennessee
One of the significant contributions by ORNL to NUCLEI Researchers is the coupled-cluster technique, which is a highly effective, systematic expansion of the nuclear wave function at a moderate cost of computation. Its solution offers detailed knowledge of the structure and disintegration of atomic nuclei and nuclear interactions. Gaute Hagen—ORNL’s lead for the NUCLEI team—is also the head for developing a flagship code NUCCOR, or NUclear Coupled Cluster Oak Ridge. NUCCOR balances between higher precision and economical computation cost.
Hagen, Gustav R. Jansen, and George Fann, at ORNL, will calculate the characteristics of nuclei and their decays. A Postdoctoral Fellow at UTK will collaborate with Papenbrock on the project. NUCLEI’s collaborators from other institutions will offer their own codes, computational techniques, as well as proficiency to the project. “Atomic nuclei exhibit very different properties as one goes from the lightest nucleus with a single nucleon—a proton—to the heaviest, consisting of about 240 nucleons [protons or neutrons],” explains Papenbrock. “In this collaboration, we have complementary methods that are good for different nuclei.”
According to Hagen, “At Oak Ridge we developed first principles methods that can describe medium mass and heavy nuclei starting from the underlying interactions between nucleons. This is remarkable progress in the field. A decade ago we were computing the structure of oxygen-16, the oxygen we breathe, which [has] 16 nucleons. Today we just submitted a paper on tin-100, which has 100 nucleons.”
The NUCLEI team will compute the characteristics of important isotopes, for example, calcium-60 including 40 neutrons and 20 protons, and hence considered to be more exceptional than calcium-40, a common stable isotope found in our teeth and bones, including 20 neutrons and 20 protons. “Calcium-60 has not been measured yet,” stated Hagen. “Nothing’s known. To go to that region—and beyond—would be a major challenge for theory. But eventually we’ll get there with the tools that we’re developing and the computing power that will be coming available to us in this SciDAC period.”
The largest nucleus proposed to be computed by the researchers from the start is lead-208. Insights on how its nucleons are held together will have a positive effect on the knowledge of superheavy elements above lead-208. In addition, the computations will be a supplement to prevalent as well as pending experiments.
The stars in ourselves
Astrophysics is a quintessentially multi-physics application. There are so many facets of physics involved; nobody can be expert in all of it. So we must build teams.
Raph Hix, ORNL, Head of TEAMS
The Researchers who are part of the TEAMS project will enhance models of the demise of massive stars, or core-collapse supernovae, which discharge chemical elements all over the galaxies, and also models of the dying hours of the lives of stars that establish the initial conditions for core-collapse supernovae. Moreover, they will also enhance models of the union of neutron stars, which form black holes and also discharge newly created elements.
Enhancing simulations of the TEAMS Researchers allows advanced microscopic nuclear physics, enhancing our knowledge of the nuclear matter states and its interactions with neutrinos. TEAMS Researchers will also investigate the aftermath of explosions that can be observed through telescopes and by applying the chemical history of our galaxy, offering observations that are comparable with simulations to corroborate models.
In the case of core-collapse supernovae, massive stars—that are 10 times the mass of Sun—form an iron core enveloped by layers of lighter elements such as carbon, silicon, helium, oxygen and hydrogen. Ultimately, the iron core gets disintegrated to form a neutron star, propelling a shock wave.
From the 1960s, Researchers have endeavored to simulate the manner in which the shock wave makes way for a supernova, by beginning with one-dimensional models that considered the star to be spherically symmetric. Simulations carried out using these models did not always culminate in explosions. In the recent past, Scientists with in-depth knowledge of the physics and faster computers ran two-dimensional (and afterward three-dimensional) core-collapse supernova models with enhanced physics.
The behavior in two or three dimensions is completely different and you get the development of big convective regions. It is neutrino energy delivered to the shock wave by convective flows that ultimately powers up the explosion. The result is an asymmetric explosion that shoots out big plumes.
Raph Hix, ORNL, Head of TEAMS
The power source stimulating such explosion is the freshly formed neutron star—where its Sun-sized mass was compressed into just 30 km—discharging enormous amount of energy quickly conveyed by neutrinos. When only a small fraction of the escaping neutrinos was captured, the shockwave is re-energized, culminating in the supernova.
The material discharged out into the galaxy by the supernova can form the next generation of stars. Elements such as oxygen which one inhales and the iron in one’s blood are discernible tracers of the chemical evolution of our galaxy from the start of the Big Bang.
“The story your atoms could tell!” exclaimed Hix. “Billions of years ago and thousands of light years away, parts of you have been through supernovae, neutron star mergers and other exotic events, and we can prove it because you carry all of the elements and isotopes that were made there. There’s a tendency when people look at the sky to say, ‘Oh, that’s the universe.’ But the universe is here too,” stated Hix, tapping his chest.
The TEAMS and NUCLEI projects and also a third SciDAC project titled Computing the Properties of Matter with Leadership Computing Resources have been supported by the DOE Office of Science. The third project will investigate the characteristics of strongly interacting particles formed of gluons and quarks. Once the findings of these projects are rendered accessible, they will be integrated with findings from the other projects and compared with experimental results to offer a more comprehensive knowledge of nuclei and nuclear interactions.