Beckoning from beyond the energy horizon like a tantalizing mirage is the ultimate power source -- hydrogen fusion. Mimicking the power that illuminates the stars, fusion harnessed to produce electric power on Earth is one of the most sophisticated technological challenges humanity has yet undertaken.
Now, after 30 years of often-frustrating effort, researchers are increasingly optomistic that control of this formidable power source is within their grasp.
In the next two years, using a device now under construction at Princeton University, physicists expect finally to demonstrate that controlled fusion is technically possible. Following this proof, it is generally expected that at least two decades will be required to engineer full-scale commercial power plants.
The justification for such a prolonged effort is the promise of abundant electric power with minimal environmental effects.
One advantage is the matter of fuel. Fusion runs on deuterium -- doubly heavy hydrogen -- which can be extracted from ordinary water. To run a 1,000 -megawatt fusion plant for one year would require only a pickup truck load of deuterium.
A second advantage is that of minimizing radioactive hazards. While fusion plants will have an inventory of volatile radioactive material, it is only a fraction of that in fission reactors. And the material involved -- tritium, a radioactive form of hydrogen -- has a half-life of 12 days as opposed to decades for much of radioactive inventory in today's reactors.
Because the "ash" of the fusion reaction is the inert gas helium, there is no radioactive waste disposal problem comparable to that in uranium-powered reactors. However, because of the intense radiation generated by the fusion reaction, materials in the reactor vessel also become radioactive and must be disposed of. This is considered a much less formidable problem than is the safe disposal of fission wastes.
Finally, there is less danger of nuclearweapons proliferation than that associated with conventional nuclear reactors. Because it does not employ uranium or plutonium, materials that can be used in nuclear weapons, fusion power as ordinarily conceived could not contribute to the spread of nuclear weapons. This does not apply to so-called "hybrid" fusion reactors -- which would breed plutonium in a surrounding blanket of uranium -- such as those being developed in the Soviet Union and advocated in the United States by the Westinghouse Electric Corporation.
However, fusion power's virtues are not easily realized. Igniting a fusion reaction -- deuterium nuclei fuse to form helium, releasing energy in the process -- requires heating a mixture of deuterium and tritium to temperatures of 50 millionto 100 million degrees C. and keeping this superhot gas together long enough for fusion to occur.
Two basic approaches to solving this problem are being explored. One uses magnetic force to form what amounts to a seamless magnetic "bottle" that holds the gas together while it is heated. The second approach involves creating a series of tiny explosions by compressing millimeter-sized pellets of deuterium by illuminating them with high energy beams of laser light or bombarding them with beams of electrically charged particles called ions.
In the magnetic confinement approach, the mainstream device now being studied is called a tokamak. Originally of Soviet design, such machines create a doughnut-shaped magnetic bottle within the main reactor vessel. The fuel, when heated to a sufficient temperature, becomes what is known as a plasma. This is a gas made up of equal numbers of positively charged particles and negatively charged electrons. While the gas as a whole is electrically neutral, the magnetic force can "grab" the electrically charged particles and confine them.
The first 30 years of fusion research were mainly concerned with keeping plasma properly within bounds. As it was heated, it tended to oscillate and squirm like a snake, breaking free of its bonds and hitting the walls of the material container where any tendency to undergo fusion was rapidly quenched.
Today, plasma instabilities appear to have been conquered as temperatures have been driven upward toward the ignition point. The major problem in recent years has been that of finding ways to drive up the temperature the last few tens of millions of degrees needed to ignite fusion. Within the last year this problem seems to have been solved. It was overcome by the development of devices that inject beams of uncharged particles into the confined plasma. In recent experiments at Lawrence Livermore Laboratory, 12 of these injectors successfully heated plasma to 130 million degrees, high enough for fusion to occur if the temperature is maintained long enough.
With these and other recent advances, it now looks as though a major milestone will be reached within two years. Experimenters are fairly confident that the tokamak fusion test reactor now under construction at Princeton will achieve what is termed "scientific breakeven." This is the point where fusion is ignited and the plasma gives off as much energy as is put into it. While significant, it is still far from the conditions required for a practical fusion power plant.
Meanwhile, the micro-explosion approach has not fared so well. Initial efforts used lasers as the "driver" to compress the pellets. This was because the scientists knew the lasers would deliver the concentrated power required.
Consequently, two large laser fusion machines were commissioned: SHIVA at Lawrence Livermore and ANTARES at Los Alamos Scientific Laboratory. "Back in the early, euphoric days we thought it would be possible to reach scientific break-even with this size [1-2 kilojoule] lasers", recalls Keith A. Brueckner of the University of California, San Diego.
But since 1978 the experts have known that their early hopes were overly optimistic. Instead of absorbing all the energy from the laser beams, the vapor created as the pellets heated reflected much of the power. As a result, when SHIVA started up last year, its performance was far below expectation. And when ANTARES goes into operation shortly, similarly disappointing results are likely, says Dr. Brueckner.
Simultaneous with the slipping fortunes of the laser approach, however, preliminary studies on using charged-particle beams have appeared increasingly encouraging, Dr. Brueckner says.
Unlike the laser case, no one can foresee problems getting the energy into the pellet once it is delivered. Beams of lightweight ions require the simplest and least expensive drivers, but a thousandfold increase in focusing them must be achieved, according to Dr Brueckner. Using beams of heavier ions would eliminate the focusing problem, but the drivers would then be more expensive.
"It would take about three years and $500 million to test the heavy ion concept," Dr. Brueckner says, adding, however, that it is the approach considered most promising.