Ever wanted to see a nuclear reactor core in action?  A new computer algorithm developed by researchers at the U.S. Department of Energy’s (DOE) Argonne National Laboratory allows scientists to view nuclear fission in much finer detail than ever before.

Development of the UNIC code is funded principally by DOE’s Office of Nuclear Energy through the Nuclear Energy Advanced Modeling and Simulation (NEAMS) program.

A team of nuclear engineers and computer scientists at Argonne National Laboratory are developing the neutron transport code UNIC, which enables researchers for the first time to obtain a highly detailed description of a nuclear reactor core.

The code could prove crucial in the development of nuclear reactors that are safe, affordable and environmentally friendly. To model the complex geometry of a reactor core requires billions of spatial elements, hundreds of angles and thousands of energy groups—all of which lead to problem sizes with quadrillions of possible solutions.

Such calculations exhaust computer memory of the largest machines, and therefore reactor modeling codes typically rely on various approximations. But approximations limit the predictive capability of computer simulations and leave considerable uncertainty in crucial reactor design and operational parameters.

“The UNIC code is intended to reduce the uncertainties and biases in reactor design calculations by progressively replacing existing multilevel averaging techniques with more direct solution methods based on explicit reactor geometries,” said Andrew Siegel, a computational scientist at Argonne and leader of Argonne’s reactor simulation group.An elevation plot of the highest energy neutron flux distributions from an axial slice of the reactor is shown superimposed over the same slice of the underlying geometry. This figure shows the rapid spatial variation in the high energy neutron distribution between within each plate along with the more slowly varying, global distribution. The figure is significant since UNIC allows researchers to capture both of these effects simultaneously.

UNIC has run successfully at DOE leadership computing facilities, home to some of the world’s fastest supercomputers, including the energy-efficient IBM Blue Gene/P at Argonne and the Cray XT5 at Oak Ridge National Laboratory. Although still under development, the code has already produced new scientific results.

In particular, the Argonne team has carried out highly detailed simulations of the Zero Power Reactor experiments on up to 163,840 processor cores of the Blue Gene/P and 222,912 processor cores of the Cray XT5, as well as on 294,912 processor cores of a Blue Gene/P at the Jülich Supercomputing Center in Germany. With UNIC, the researchers have successfully represented the details of the full reactor geometry for the first time and have been able to compare the results directly with the experimental data.

Argonne’s UNIC code provides a powerful new tool for designers of safe, environmentally friendly nuclear reactors – a key component of our nation’s current and future energy needs. By integrating innovative design features with state-of-the-art numerical solvers, UNIC allows researchers not only to better understand the behavior of existing reactor systems but also to predict the behavior of many of the newly proposed systems having untested design characteristics.

Development of the UNIC code is funded principally by DOE’s Office of Nuclear Energy through the Nuclear Energy Advanced Modeling and Simulation (NEAMS) program. The Argonne UNIC project is a key part of the NEAMS efforts to replace the traditional “test-based” approach to nuclear systems design with a new “science-based” approach in which advanced modeling and simulation play a dominant role.

A video of a more detailed simulation of the Zero Power Reactor experiment is available here

Advanced supercomputer simulations could be a big hit for truckers and the people who design guardrails, protective barriers and roadway signs. Srdjan Simunovic of Oak Ridge National Laboratory is part of a team conducting a study aimed at gaining a better understanding of crash performance of these safety structures.

"Very limited computational simulation work has been conducted on crash performance of barriers when impacted by medium- and heavy-duty trucks because of the computational cost and complexity of full-scale truck models," Simunovic said.

The design and engineering of these structures strongly influence the injury-causing g-forces experienced by vehicle occupants and whether the trucks are redirected back into traffic, causing additional hazards.

Partners in the study, funded by the National Transportation Research Center Inc., are Battelle Memorial Institute and the University of Tennessee.

Daniel Pedro knew when he was a sophomore at Santa Fe Indian School that he wanted to be an anthropologist. He also knew that as a Zuni, he would not be able to touch human remains – a common task for physical anthropologists.

“It was kind of a barrier,” said Pedro, a 20-year-old freshman at the University of New Mexico-Gallup. “I had to find a way to work around it.”

Pedro began to look for that way through his participation in the New Mexico Supercomputing Challenge. The Challenge aims to teach teams of middle and high schools students how to use powerful computers to analyze, model and solve real world problems and awards prizes in various categories.

Pedro hit on the idea of studying the faces of living puebloans in search of consistent similarities and then projecting that data onto the past as a way to identify and repatriate skeletal remains. As stated in the executive summary of his project, “My goal … is to make it easier for anthropologists to figure out which tribe/pueblo the remains belong to on the computer, instead of disrespecting Native customs and damaging the skull.”

An early advisor, UNM Curator of Human Osteology Heather Edgar, told Pedro that the people of the pueblos, both present and past, were too mixed to make the sort of determinations he was seeking. Nevertheless, she was impressed by his inventive approach to problem solving, and encouraged him by giving advice on how to go about his project. She also gave him a medical diagram of a human skull with which to start his studies.

“We need a Native perspective in anthropology, and especially a perspective that comes from working with living communities,” Edgar said.

Pedro’s unique project soon attracted several other advisors and mentors.

“They were impressed by the fact it was a student who wanted to do this kind of work, and a high school student and a Native American at that,” Pedro said.

Pedro began to work in a computer program called StarLogo, which allowed him to rotate two objects side by side and compare the objects in different profiles.  He had decided to concentrate on the human skull, comparing shapes that represented skulls. His goal was to create a method for anthropologists to determine which tribe or pueblo a skull might belong to with only minimal handling. The result was an entry for the Supercomputing Challenge called “Scan of the Past.”

“He learned a lot about the mathematics of 3D computer graphics and the rotation and scaling of 3D objects on the computer,” said Irene Lee, who oversees a grant program at Santa Fe Institute, and was previously lead facilitator for the SFI-MIT Adventures in Modeling program in Santa Fe when she worked with Pedro.

For this phase of his project, he received the Judges’ Choice Award for “Integrating Computation into Anthropology” from the Supercomputing Challenge.

The second phase of his work was on a new version of the “Scan of the Past,” with the help of Steve Guerin of Redfish Group, a Santa Fe-based business that specializes in data mining and visualizing. Guerin helped Pedro during his senior year construct a proxy data set, which would allow him to practice clustering techniques and classification algorithms, or in other words, construct real world data. Pedro learned how to integrate actual facial data collected after he photographed and studied 15 landmarks on the faces of 45 individuals – fellow students whom he persuaded to participate in his project. Generally, says Edgar, studies are made with as many as 50 landmarks on a human skull, but because Pedro was concentrated on faces, his study was limited to far fewer.

Although this phase of the project did not earn an award, he did receive an award from the Supercomputing Challenge in 2008-2009 for creating a graphic poster and creating a logo.

“Daniel took on a computational challenge that was meaningful to him and his community,” Lee said. “He is a great role model of a self-directed student researcher. He found an interesting, unsolved problem he could address. He overcame many obstacles and persevered with the project over several years.”

“It was great to have help from so many mentors,” Pedro said. “I had wondered if my project would be taken seriously because this was something really new.”

After graduating from high school in 2008, Daniel went on to enroll in UNM-Gallup, where he is studying, among other subjects, anthropology with Teresa Wilkins, professor of anthropology. Last year, he got a taste of the museum work he hopes to make a career by working at A:shiwi A:wan Museum and Heritage Center in Zuni, where he learned how to care for exhibits and worked with the photo collection. He also got some good career experience this past summer by participating in the Conference on Archaeoastronomy of the American Southwest, Camp Verde, Ariz., where, with researcher Anna Sofaer, writer, artist and founder of the Solstice Project, he presented a poster on a new interactive computer model of the Chaco Canyon Sun Dagger site.

His work with Sofaer helped him see that Native Americans “did marvelous things,” and reinforced his idea that, when studying historic sites, “It’s best to listen to Native American oral traditions about what happened at these sites. If we can integrate these traditions with what we can learn from modern technology, we can create another level of thinking.”

As Pedro continues his journey toward a bachelor’s degree in Southwest Studies at UNMG, and beyond, to a Ph.D., the intention that inspired his high school project will be very much with him. He wants to continue to explore ways to use technology to repatriate human remains and relics to the tribes they belong to. At the same time, he wants to build on what he learned from his high school project and his work with the Solstice Project to bring computers into anthropological work in a way that will help Native Americans understand who they are.

“As an example, time may overtake the original Sun Dagger site and it will become part of nature, but a replica of the model will be there to teach how Native Americans used the solstice at Chaco, and how they measured time,” Pedro said.

Wilkins applauds Native students like Pedro who are looking to apply “sophisticated technology to [solve] real problems,” and echoes his hope that today’s Natives will become empowered to make their own identifications of remains in order to repatriate them. She also believes that such an applied approach to anthropology as Daniel Pedro’s project undertook may be “highly significant in empowering Native people to conduct their own research.”

Pedro also hopes his vocation as anthropologist will help show Native Americans that eventually, they should not have to take classes to “be native.” After all, he points out, most of the non-natives who have taught American Indians about their history and culture cannot have complete information because, “There is a limit as to how much we can share. Keeping the culture or religion with the community keeps our identity within the community, rather than having it spill out.” Ideally, he says, those studying and interpreting the research some day will be Natives who will not only share this knowledge with their communities, but also mediate what is shared with non-Natives.

Berkeley scientists shine new light on green plant secrets

The future of clean green solar power may well hinge on scientists being able to unravel the mysteries of photosynthesis, the process by which green plants convert sunlight into electrochemical energy. To this end, researchers with the U.S. Department of Energy (DOE)'s Lawrence Berkeley National Laboratory (Berkeley Lab) and the University of California (UC), Berkeley have recorded the first observation and characterization of a critical physical phenomenon behind photosynthesis known as quantum entanglement.

Previous experiments led by Graham Fleming, a physical chemist holding joint appointments with Berkeley Lab and UC Berkeley, pointed to quantum mechanical effects as the key to the ability of green plants, through photosynthesis, to almost instantaneously transfer solar energy from molecules in light harvesting complexes to molecules in electrochemical reaction centers. Now a new collaborative team that includes Fleming have identified entanglement as a natural feature of these quantum effects. When two quantum-sized particles, for example a pair of electrons, are "entangled," any change to one will be instantly reflected in the other, no matter how far apart they might be. Though physically separated, the two particles act as a single entity.Mohan Sarovar (seated) and (from left) Akihito Ishizaki, Birgitta Whaley and Graham Fleming carried out the first observation and characterization of quantum entanglement in a real biological system.  Credit: Photo by Roy Kaltschmidt, Berkeley Lab Public Affairs

"This is the first study to show that entanglement, perhaps the most distinctive property of quantum mechanical systems, is present across an entire light harvesting complex," says Mohan Sarovar, a post-doctoral researcher under UC Berkeley chemistry professor Birgitta Whaley at the Berkeley Center for Quantum Information and Computation. "While there have been prior investigations of entanglement in toy systems that were motivated by biology, this is the first instance in which entanglement has been examined and quantified in a real biological system."

The results of this study hold implications not only for the development of artificial photosynthesis systems as a renewable non-polluting source of electrical energy, but also for the future development of quantum-based technologies in areas such as supercomputing - a quantum computer could perform certain operations thousands of times faster than any conventional computer.

"The lessons we're learning about the quantum aspects of light harvesting in natural systems can be applied to the design of artificial photosynthetic systems that are even better," Sarovar says. "The organic structures in light harvesting complexes and their synthetic mimics could also serve as useful components of quantum computers or other quantum-enhanced devices, such as wires for the transfer of information."

What may prove to be this study's most significant revelation is that contrary to the popular scientific notion that entanglement is a fragile and exotic property, difficult to engineer and maintain, the Berkeley researchers have demonstrated that entanglement can exist and persist in the chaotic chemical complexity of a biological system.

"We present strong evidence for quantum entanglement in noisy non-equilibrium systems at high temperatures by determining the timescales and temperatures for which entanglement is observable in a protein structure that is central to photosynthesis in certain bacteria," Sarovar says.

Sarovar is a co-author with Fleming and Whaley of a paper describing this research that appears on-line in the journal Nature Physics titled "Quantum entanglement in photosynthetic light-harvesting complexes." Also co-authoring this paper was Akihito Ishizaki in Fleming's research group.

Green plants and certain bacteria are able to transfer the energy harvested from sunlight through a network of light harvesting pigment-protein complexes and into reaction centers with nearly 100-percent efficiency. Speed is the key – the transfer of the solar energy takes place so fast that little energy is wasted as heat. In 2007, Fleming and his research group reported the first direct evidence that this essentially instantaneous energy transfer was made possible by a remarkably long-lived, wavelike electronic quantum coherence.

Using electronic spectroscopy measurements made on a femtosecond (millionths of a billionth of a second) time-scale, Fleming and his group discovered the existence of "quantum beating" signals, coherent electronic oscillations in both donor and acceptor molecules. These oscillations are generated by the excitation energy from captured solar photons, like the waves formed when stones are tossed into a pond. The wavelike quality of the oscillations enables them to simultaneously sample all the potential energy transfer pathways in the photosynthetic system and choose the most efficient. Subsequent studies by Fleming and his group identified a closely packed pigment-protein complex in the light harvesting portion of the photosynthetic system as the source of coherent oscillations.

"Our results suggested that correlated protein environments surrounding pigment molecules (such as chlorophyll) preserve quantum coherence in photosynthetic complexes, allowing the excitation energy to move coherently in space, which in turn enables highly efficient energy harvesting and trapping in photosynthesis," Fleming says.

In this new study, a reliable model of light harvesting dynamics developed by Ishizaki and Fleming was combined with the quantum information research of Whaley and Sarovar to show that quantum entanglement emerges as the quantum coherence in photosynthesis systems evolves. The focus of their study was the Fenna-Matthews-Olson (FMO) photosynthetic light-harvesting protein, a molecular complex found in green sulfur bacteria that is considered a model system for studying photosynthetic energy transfer because it consists of only seven pigment molecules whose chemistry has been well characterized.

"We found numerical evidence for the existence of entanglement in the FMO complex that persisted over picosecond timescales, essentially until the excitation energy was trapped by the reaction center," Sarovar says.

"This is remarkable in a biological or disordered system at physiological temperatures, and illustrates that non-equilibrium multipartite entanglement can exist for relatively long times, even in highly decoherent environments."

The research team also found that entanglement persisted across distances of about 30 angstroms (one angstrom is the diameter of a hydrogen atom), but this length-scale was viewed as a product of the relatively small size of the FMO complex, rather than a limitation of the effect itself.

"We expect that long-lived, non-equilibrium entanglement will also be present in larger light harvesting antenna complexes, such as LH1 and LH2, and that in such larger light harvesting complexes it may also be possible to create and support multiple excitations in order to access a richer variety of entangled states," says Sarovar.

The research team was surprised to see that significant entanglement persisted between molecules in the light harvesting complex that were not strongly coupled (connected) through their electronic and vibrational states. They were also surprised to see how little impact temperature had on the degree of entanglement.

"In the field of quantum information, temperature is usually considered very deleterious to quantum properties such as entanglement," Sarovar says. "But in systems such as light harvesting complexes, we see that entanglement can be relatively immune to the effects of increased temperature."

 

Over the last six years Enabling Grids for E-sciencE (EGEE) has grown to become the world's largest multi-disciplinary grid infrastructure, with tens of thousands of users. In Barcelona this month the flagship EC-funded project meets for its final annual conference. Next year will see the inauguration of EGEE's successor, the European Grid Initiative (EGI), and this pivotal gathering of the grid community will allow for reflection on the successes of the project, as well as marking a major step forward on the path towards a nationally-focused sustainable grid infrastructure, to benefit all European researchers for years to come. The transition from EGEE to EGI represents a welcome move from short-term project funding to sustained support on a national and international level, which will enable users to continue using grid infrastructures – now and in the future.

Successes have been numerous and EGEE already caters for a multitude of disciplines. This final annual conference is an opportunity for the user communities to showcase their work and achievements, and demonstrate the power of grid technology. There will be live demonstrations from diverse scientific research fields, including many from medical research - notably neuGrid, studying degenerative diseases such as Alzheimer's, RadioTherapy Grid optimising the use of radiotherapy in cancer treatment and EUAsiaGrid helping research into monitoring future flu pandemics.

While the medical community has benefited greatly it is not the only one to increasingly rely on grids. The computing power and data storage offered by EGEE has allowed scientists to think big - from modelling weather using the grid with WRF4G, predicting the distribution of marine life with AquaMaps, to EUAsiaGrid’s software helping authorities to cope in the aftermath of an earthquake. The breadth and quality of scientific research supported by EGEE is a testament to the versatility of computing grids.

It is this versatility that has attracted industry. EGEE has worked with the business community since the early days of grids. Each year, the dedicated business track increases the collaboration between EGEE and industry.  This year is no different, demonstrating the evolution of grid computing by highlighting not only key areas such as cloud computing and service level agreements, but also unveiling prime examples of technology transfer from science and academia to commercialisation.

With cloud computing generating increasing excitement in the business community, EGEE is working to integrate the two services. Two sessions at EGEE’09 will present a range of projects working on grids and clouds, such as StratusLab which is bringing grids and clouds together to create benefits for both science and industry.

As in previous years, the conference has attracted major grid computing figures from around the world as keynote speakers, including locally-based speaker Gonzalo Merino from PIC, who is in charge of Spain’s Tier-1 data centre for the Large Hadron Collider, Jennifer Schopf, from the National Science Foundation’s Office of CyberInfrastructure in the US and Kostas Glinos, who leads the Géant & e-Infrastructures Unit of the Directorate General for Information Society and Media at the European Commission. EGEE’09 will also be a key milestone in the preparations of pan-European projects defined within the European Strategy Forum on Research Infrastructures (ESFRI), presented at the conference by Professor John Wood, chair of the European Research Area Board and former Chair of ESFRI.

Page 10 of 26