The race to net-energy-gain fusion

Pretty picture: A computer-generated cutaway image of JET during a plasma experiment. Image courtesy EFDA JET.

Nuclear fusion has long been touted as a panacea for the world's current energy crisis. However, reactions which produce more energy than is put in are still some way off. There are a number of competing methods for generating fusion currently being explored by scientists across the globe, but which one will achieve the milestone of breaking even in terms of energy input versus energy output first?

There is currently little consensus among scientists, but perhaps computer models can throw some light on the race to become the world's new energy savior. There are at present two front-runners in this race: the first technique is called magnetic confinement. This is typically done using a magnetic bottle called a tokamak to contain hot fusion fuel. The second technique is called inertial confinement and is the localized compression of a small amount of targeted fuel over nanoseconds to create fusion. This can be done using ultra-intense lasers or a magnetic pulse.

The Joint European Torus (JET) project, in the Culham Centre for Fusion Energy, UK, is the world’s largest experimental nuclear fusion facility that uses a magnetic tokamak. Inside its tokamak, magnetic fields fuse two hydrogen isotopes, deuterium and tritium. To achieve this, the fuel is heated to very high temperatures in which ion and electrons separate into a state of matter:  plasma. The experiments are supported by a host of supercomputers, which are used for data analysis and storage, engineering design, materials research, plasma simulation and real-time reactor control.

JET required 24 megawatts of power input to produce 16 megawatts of power output in 1997, during the last big deuterium and tritium experiment. The reactor has been running since last September, with a new inner wall. But, if anyone hopes that it will produce a net-energy-gain fusion they’ll be disappointed; this is the goal of its successor, the International Thermonuclear Experimental Reactor (ITER), under construction in Cadarache, France, which will be a major step towards commercial fusion power plants.

Don't switch it on please: NIF Target Chamber: Interior of NIF target chamber. A service module carrying technicians is on the left. The target positioner, which holds the target, is on the right. Image courtesy NIF.

“Last year, we carried out the first experiments with the new ITER-like wall. Doing this in just deuterium, rather than in deuterium-tritium mixtures, allows us to test most of the relevant physics in a more flexible manner,” says Lorne Horton, head of the JET department at the European Fusion Development Agreement.

The new wall at JET  is a combination of beryllium and tungsten armor, which has demonstrated the hoped-for greater-than-10-fold reduction in trapping deuterium, thus giving efficient fusion results. Further tests for ITER will continue at JET into 2015 and 2016. 

 Big, big, lasers

Fast-ignition reaction fusion is a variation of inertial confinement. This is the use of powerful lasers to compress the fuel, then a second laserpulse heats and starts fusion ignition. In July 2012, the Lawrence Livermore National Laboratory's National Ignition Facility (NIF) fired its 192 laser beams, within a few trillionths of a second of each other, with more than 500 trillion watts (terawatts or TW) of power and a total energy of 1.85 megajoules of ultraviolet laser light to a target. Five hundred terawatts is 1,000 times more power than the entire US uses at any instant in time.

When a laser this intense hits matter, such as a tiny condensed pellet of hydrogen, the hydrogen atoms fuse to form a plasma of helium, releasing energy. It’s similar to a car’s gasoline engine, which uses a spark to ignite fuel that is only partly compressed, which enables faster ignition and a more efficient energy release. But, even though computer models had shown that the NIF were on target for their US congressional deadline of net-energy-gain fusion by 30 September 2012, this proved not to be the case.

Computer models predict results way beyond break-even fusion

Recently, and for the first time, the latest computer simulations from the Partnership for Advanced Computing in Europe (PRACE) supercomputer collaboration modeled a full-scale interaction of ignition lasers with compressed fuel for inertial fusion. Inertial fusion refers to the localized compression of a small amount of targeted fuel over nanoseconds. “It has provided critical and more reliable information on the physics and parameters for multi-scale models that are being used to design future experiments,” says project leader Luís Silva, professor of physics at the Instituto Superior Técnico, Lisbon, Portugal.

Zap! Sandia's Z-Machine concentrates electrical energy and turns it into short pulses of enormous power, which are used to generate X-rays and gamma rays. High-gain nuclear fusion could be achieved in a pre-heated cylindrical container immersed in strong magnetic fields: Magnetized Inertial Fusion (MagLIF). The cylinder, known as a liner, is connected to the Z machine, which delivers 26 million amps in a pulse. This  current passing down the walls of the cylinder creates a magnetic field instantly crushing it, and compressing and heating the fusion fuel. Image courtesy Sandia.

“We had previously done smaller scale simulations with smaller targets. The breakthrough is that we are not modeling an idealized situation, but a real fast-ignition scenario,” says Silva in the latest PRACE newsletter. This means that fast ignition fusion can, if accurately predicted, work better in targets of realistic size. The researchers used a unique numerical code called OSIRIS, that ran for 30 million CPU hours on the Jugene Blue Gene/P supercomputer to simulate the evolution of hundreds of millions of particles.

“We are exploring an alternative concept that, in principle, will require less laser energy to achieve ignition. The fact that the more ‘brute force’ approach at NIF has not yet demonstrated ignition is a clear indication that we need more studies and more physical understanding of the physics, which is what is being pursued at NIF.

But, we also need to explore alternative concepts. We are confident that the methodology employed and the results obtained in our work will be of significant importance in determining which concepts will be feasible and worth exploring in the future,” says Silva.  

Going back to the US, the Sandia National Laboratories stated earlier this year that their computer simulations showed a predicted output energy that was many times greater than the energy fed in, but using a magnetic pulse instead of lasers, for inertial confinement fusion.

Their simulations showed about 100 times the output of 60 million amperes (MA) input, rising to 1,000 times the incoming pulse of 70 MA, and 50 times more efficient than using X-rays, which was Sandia’s previous method. Their research paper is published in Physical Review Letters.

Sandia plans to produce reliable fusion-made electricity from seawater. Currently, they’re testing physical equipment and comparing it with these computer simulations. But, simulations are flawed at some level and even as they’re improved they’ll continue to remain flawed according to lead computer simulation researcher Steve Slutz of Sandia National Laboratories.

“The scientific method depends on both experiment and theory and they motivate each other. Simulations have become very sophisticated, as the power of computers has developed," says Steve Slutz. "Presently simulation codes can adequately predict many physical experiments. However, nature is even more complex and some experiments cannot presently be accurately predicted, but this does not mean that simulations are of no use in these situations. Numerical simulations still provide our best estimates of what to expect in complex experiments and can be used to design future experiments. Complex experiments requiring numerical simulations need to be taken in steps, but the size of those steps will probably always be the subject of debate.”

Their latest research paper, published on 28 September 2012 in Physical Review Papers, states that Sandia have passed their first three tests and are on track to achieve break-even fusion. Laboratory fusion results are expected by late 2013.