GAITHERSBURG, Md., Nov. 25 (UPI) -- Researchers must integrate dozens of computer codes to simulate the components of a nuclear fusion reaction, as well as update supercomputer hardware, in order to help humanity harness the near-limitless power source, a government panel reported Monday.
The Department of Energy's Fusion Energy Sciences Advisory Committee endorsed a report calling for a 15-year Fusion Simulation Project to build a system capable of modeling a complete fusion reactor.
At present, U.S. researchers deal with more than 50 separate computer programs in designing and analyzing the physics underlying the process, said Jill Dahlburg, director of Inertial Fusion Technology at General Atomics in San Diego.
The codes, clustered in four main areas, cover differing timescales within a fusion reaction, said Dahlburg, who chaired FESAC's Integrated Simulation and Optimization of Fusion Systems Subcommittee.
"If you say let's integrate all the codes over all the timescales, there's really no roadmap for doing that," Dahlburg told the committee. "Our approach deals with this in (pairs of problems) instead."
The idea of fusion reactors has been studied for more than 40 years. If achieved, the concept would provide a nearly inexhaustible power source. As two hydrogen atoms fuse into a helium atom, they release great quantities of energy while avoiding the greenhouse-gas emissions of fossil fuels and much less radioactive byproduct than nuclear fission.
The catch in harnessing a fusion reaction lies in controlling the prerequisite conditions -- temperature and pressure levels that reduce matter to a cloud of atomic nuclei and electrons called plasma, far beyond what physical materials can stand.
Current research uses magnetic fields to confine the reaction. Some of the codes help predict how the plasma will "ignite" and move inside the magnetic "bottle." Others deal with how the bottle's walls could cool or otherwise disrupt the plasma.
Existing DOE programs hold the core of expertise needed to integrate the codes, Dahlburg said, but new researchers must be attracted to the project for it to succeed. The FSP's first 5 years, therefore, should be funded at $20 million per year to convince talented physicists and computer science experts to come on board, she said. The overall 15-year plan would cost about $400 million, she said.
Initial funding levels might be inadequate, however, said Glen Wurden, a team leader in magnetic fusion experiments at Los Alamos National Laboratory. The problems involved are at the same order of complexity as effectively simulating nuclear weapons for maintaining the nation's stockpile, he told the committee during a public comment period, and that program require much more than $20 million a year.
The pace of hardware improvements should give FSP the computer horsepower to effectively run the integrated codes, said James Corones, president of the Krell Institute, a technology think tank. Japan's Earth Simulator, the world's current supercomputing champ, is the sort of system that could accurately simulate the interior of a magnetic bottle, said Corones, a member of the ISOFS committee. A full-reactor simulator would need to be about 1,000 times faster and manipulate about 1,000 times more data than the Earth Simulator, but such advances are expected during the FSP's lifetime, he said.
FESAC is also examining possible fusion applications beyond power generation, said committee member Kathryn McCarthy, manager of the Nuclear Engineering Design and Research Department at the Idaho National Engineering and Environmental Laboratory. For example, the heat generated by a fusion reactor, combined with plentiful electricity during off-peak hours, could help generate hydrogen for emissions-free fuel-cell vehicles, she told United Press International.
Fusion also could help deal with waste products from current nuclear power plants, McCarthy said. The reaction generates large quantities of very energetic subatomic particles called neutrons, which could be directed at radioactive elements to induce them to become inert more quickly, a process called transmutation, she said.