When most people think about the as yet unrealized promise of thermonuclear fusion as a limitless energy source, they tend to think about Big Science, in the form of giant lasers blasting tiny targets of hydrogen, or of equally large Tokamak reactors built with giant magnets. They also tend to think of fusion as a pipe-dream technology that is a long way off.
Recently a small, underfunded startup company in New Jersey, Lawrenceville Plasma Physics (LPP) has made significant progress using a technology that is orders of magnitude smaller, cheaper, and simpler than the Big Science approaches: the dense plasma focus, or DPF.
Educational stuff, if you don't know much about fusion
Thermonuclear fusion is the process of taking small atoms, like hydrogen, and fusing their nuclei together to make bigger atoms, like helium, and releasing a lot of energy in the process. This is not at all easy to do.
First, atoms are usually surrounded by electron shells that repel other atoms. There is a fairly easy way around this barrier, however, because electrons are easy to manipulate. You strip away all the electrons from your atoms with a big electric current, and the remaining atomic nuclei are electrically charged ions. If you do that to the atoms in a gas, the ionized gas forms a new state of matter called a plasma, which is not usually seen on Earth (except in neon signs), and seldom for very long – like lightning, the most common form of plasma on Earth. Unlike normal matter, plasma is both electrically charged and electrically conductive. Electricity can flow through plasma just like it flows through metal.
Forming a plasma is the easy part. The problem is that those bare atomic nuclei are all positively charged – which means they repel each other, and the closer they get to each other, the greater the repulsion. But in order to achieve fusion, those nuclei have to get very, very close together: so close that the strong nuclear force is great enough to overcome the electrical repulsion. If that happens, the nuclei will fuse, and release a Whole Lot of Energy when they do.
The way to get those nuclei close together is to get the plasma very, very hot. Which simply means that the nuclei in the plasma are moving very, very fast – so fast that their kinetic energy will sometimes be great enough to overcome the electrical repulsion of the atomic nuclei. But the temperatures needed for fusion are so high that no physical substance is capable of containing a plasma that hot.
So the big, big issue in fusion research is the problem known as confinement: how do you confine very hot plasma at a pressure great enough to cause them to fuse? There are two basic approaches: inertial confinement and magnetic confinement.
Inertial confinement is the approach that uses large lasers to blast tiny little pellets of hydrogen, or other fusible material. If you can put enough energy into a small enough volume of space in a short enough time, you should be able to get atoms to (a) ionize, and (b) fuse, before they have time to move away from each other. In other words, the inertia of the atoms themselves would prevent them from moving fast enough to get away from all the other atoms in the pellet. It takes very, very large lasers for this idea to work.
Magnetic confinement uses magnetic fields, instead of physical substances, to make a magnetic "bottle" to contain the hot plasma. There are various geometries for magnetic bottles that have been tried, and are still being tried. Among these geometries are the the Tokamak, Spheromak, and Stellerator.
There are also hybrid devices, like Sandia Labs' Z machine; it uses magnetic confinement on one plasma to generate a lot of x-rays, and uses those x-rays and inertial confinement to try to fuse a second plasma.
All of these devices are capable of fusing atoms on a small scale, and are doing so today. But none of them is close to the break-even point, which is the point at which the energy produced in fusion is larger than the energy used in running the machine itself.
The Dense Plasma Focus
The dense plasma focus, or DPF, is a gnat among leviathans in fusion research. The DPF was invented (twice: independently in Russia and the US) in the late 1950's, and can be made easily and cheaply enough that many university physics departments have them. They are often used as a means to generate x-rays, because hot plasmas usually produce x-rays as a byproduct.
Unlike the giant machines used for intertial confinement and magnetic confinement, the guts of a DPF machine could fit easily on a table top. The DPF works by running a very strong electrical current through the plasma, in a toroidal geometry. The strong current produces a strong magnetic field, and that, combined with the geometry of the machine, causes the plasma to "pinch", which is to say, it forms a plasmoid: an unstable knot of plasma filaments, just a few dozen microns in diameter. The internal magnetic field in the plasma itself causes the plasmoid to collapse in on itself, briefly creating very high temperatures and pressures before it destroys itself.
It had long been thought that the DPF could never be used for fusion, because too much of the energy in the plasmoid gets radiated away as x-rays before fusion could occur. But a few years ago, Dr. Eric Lerner (currently with LPP) determined theoretically that under certain conditions, the plasmoid could be made too dense for x-rays to be easily created, leaving enough of the thermal energy intact for fusion to occur at break-even energies. According to Dr. Lerner's calculations, however, it would take a much larger current than had ever before been used in a DPF device to achieve those conditions.
With no budget and no backing, but a belief in his math, Dr. Lerner became a wandering salesman for DPF as a simple and cheap fusion device. In late 2008, not long after he gave one of Google's famous "Tech Talks", he secured private funding to build his machine. Last fall the DPF machine at LPP, dubbed "Focus Fusion 1" or FF1, achieved its first pinch on budget and ahead of schedule.
Recent Developments
Like any experimental device, FF1 has had annoying glitches. The currents LPP is aiming for are in excess of two million amps, at tens of thousands of volts. That energy is stored in a capacitor bank, and the machine has to release all of it within about 25 nanoseconds. But there is no switch in existence that can handle that much current. So FF1 uses a set of 12 plasma switches, all of which are supposed to fire at the same time. Usually they don't, which means that single shots of the device tend to be at lower currents than designed.
(A plasma switch is a cylinder containing an easily ionized gas, with electrodes at either end. A spark plug in the cylinder ionizes the gas, changing it from an insulator to a conductor, thus throwing the switch.)
Also, each switch is triggered by a standard automotive spark plug, and the very high currents have been melting the nickel electrodes of the plugs, requiring frequent replacement. The team has just received a new alloy of tungsten and rhodium, from which they hope to fashion new, melt-free electrodes for all the plugs.
In spite of the problems, notable progress has been made. In February, an experiment that added angular momentum to the plasmoid was successful, showing that a rotating plasmoid can produce 8 to 10 times the number of neutrons (fusion reactions) as a non-rotating plasmoid.
Using deuterium fuel, instrumentation on FF1 has shown high ion energy in the plasmoid of 40 to 65 KeV (equivalent to 440 to 715 million degrees). These are among the highest, if not the highest, ion temperatures ever achieved in a DPF device, in spite of the fact that currents have been lower than expected. Neutrons produced in fusion reactions during these shots have been measured within energy ranges expected for fusion reactions. (I should note here that deuterium is being used only for experimental purposes. The ultimate goal is to use borane gas [BxHy] as fuel, which will ionize into bare protons [from the Hydrogen] and Boron-11 nuclei, which in turn fuse to form 3 Helium nuclei. This is known as the p-B11 reaction, and it's very useful in power production because it produces no neutrons, hence no radioactive waste.)
On April 26, FF1 had a breakthrough of sorts when all 12 switches fired simultaneously for four consecutive shots. These shots were done with the fill gas in the machine at a relatively high pressure of 40 torr, which was done deliberately to prevent a pinch from occurring. However, with all 12 switches activated the pinch occurred anyway at the tail end of the current pulse, with currents measured above 1 megamp (1 million amps). This is the largest current ever achieved in a DPF device, and the highest gas fill pressure ever to achieve plasmoid pinch. Once FF1 is working at its maximum design current of 2.4 megamps, a 40 torr fill gas pressure should be about optimal, so it is encouraging indeed to see pinches already occurring there.
And finally, LPP has been able to stably simulate a plasma filament for the first time, using a new algorithm – but in one dimension only: radially from the center of the filament. (Odd that a plasma filament, which we see every time we see a lightning bolt, has never been adequately simulated by physicists until now.) The next step will be to extend the simulation to two dimensions, and ultimately three. It is hoped that these simulations will yield new understandings of plasma physics, including new insights into how fusion reactions can be optimized.