THE boxy, galvanized-steel building at the Lawrence Livermore National Laboratory (LLNL) is an unlikely looking workshop for ''cathedral'' builders. Yet this unprepossessing building is part of modern physics' most Promethean dream - the attempt to duplicate a star's fire to provide power for the 21st century.
''The fusion program is really remarkable in our society. It is like the gothic cathedrals that took a century or more to construct, in that people committed resources to something they will never see completed. It is an investment for future generations,'' says Harry Dreicer, director of the fusion program at Los Alamos National Laboratory.
rr Five stories high and the size of a football field, Lawrence Livermore's contribution to the ''cathedral'' is filled with scaffolding and large, stainless-steel cylinders. In an adjacent lot, massive, twisted 30-foot stainless steel magnets stand like a modern sculpture. These magnets will generate part of the magnetic bottle that will hold a sausage of electrically charged gas called a plasma. The plasma, with little more substance than a vacuum, will be heated to tens of millions of degrees.
When a mixture of two varieties (isotopes) of hydrogen gas is heated to a high enough temperature and maintained at sufficient pressure for a long enough time, scientists say they can trigger a fusion reaction. In this reaction, atoms of hydrogen are transmuted into helium. And in the process, small amounts of mass are converted into large amounts of energy.
The relativistic conversion of mass into energy is also at the heart of nuclear fission, the process that powers today's nuclear reactors. Fission and fusion bear the same relationship as the atomic and hydrogen bomb. The fission chain reaction involves splitting the heaviest of atoms: uranium, thorium, and plutonium. This is easier to achieve than fusion but releases less energy and creates more radioactivity. Fusion is cleaner and it unleashes tremendous amounts of energy. But it needs stellar conditions to ignite. Thus, all thermonuclear warheads are detonated by atomic bombs.
For more than 30 years several thousand scientists in the United States, the Soviet Union, Europe, and Japan have been trying to recreate the conditions required to ignite this reaction under controlled conditions. Earlier this year, researchers at the Massachusetts Institute of Technology reached this regime for the first time. But because their machine lacks adequate radiation shielding, the scientists did not use the correct hydrogen mixture for ignition.
''While this was only twice what had been achieved before, it passed a magic number and we all breathed a sigh of relief,'' says Stephen O. Dean, president of the Fusion Power Associates, a nonprofit group incorporated to aid in the commercialization of fusion technology.
The first machine expected to get as much energy out of the fusion reaction as is required to create it is the Tokamak Fusion Test Reactor (TFTR) at Princeton University. Tokamak is a combination of Russian words meaning a toroidal magnetic chamber. This mid-1960s Soviet design, which holds plasma in a doughnut-shaped magnetic bottle, was the key to the current progress toward the demonstration that controlled fusion is scientifically possible.
To get this far has taken a world effort and more than $14 billion. Gerald L. Kulcinski, a professor of nuclear engineering at the University of Wisconsin who has headed a number of fusion-reactor design studies, estimates that $50 billion will be spent before fusion power contributes a single watt to the world's electricity supply.
''These are sobering figures. They make you stop and ask, 'What is the purpose of this investment?' '' he says.
THE cathedral analogy has also been adopted by fusion's critics. Commercial design studies have generally concluded that, even when scaled two to five times larger than current fission reactors, the cost of electricity they might produce would be twice the cost of that produced from current sources. This has led critics to characterize fusion reactors as useless ''superconducting cathedrals, '' superconducting because such machines require magnets made of material that, when cooled to under 200 degrees Celsius below zero, loses nearly all of its electrical resistance.
The fact that tokamak-type devices appear unlikely ever to produce energy at competitive prices has not greatly discouraged the scientists involved. They feel that with increased knowledge of the fusion process, it should prove possible to create an energy technology fundamentally superior to any the world has experienced.
The reasons for this optimism are the intrinsic qualities of the fusion process. The source of one type of hydrogen fuel, deuterium, is seawater. The fuel, tritium, can be transmuted from the common metal, lithium, in the reactor itself.
Although the fusion reaction emits radiation, large amounts of radioactive materials are not intrinsic to the process. According to Charles C. Baker, director of the Argonne National Laboratory's fusion efforts, such a reactor built of current materials would contain an inventory of radioactive material 100 times less biologically hazardous than a typical fission reactor. And, with further development, this could be reduced by 10 to 100 times more.
Finally, the ''ash'' of fusion is helium, an inoffensive gas. It does not produce carbon dioxide, a gas created by burning wood and fossil fuels, which is building up in the atmosphere. Climatologists believe CO2 is subtly altering the climate by driving up the average world temperature. Balanced against these potential advantages, however, are engineering challenges every bit as formidable as the scientific obstacles that have been overcome.
''We still have an enormous task ahead of us. Engineering feasibility may be even harder to achieve than scientific feasibility. It is likely to be much harder than many imagine,'' says Harold K. Forsen, of the construction company Bechtel National.
The fusion effort remains dominated by physicists, although in recent years a number of nuclear engineers have deserted fission for fusion. The various world programs have single-mindedly focused on achieving ignition. As this goal comes ever nearer, concern is gradually shifting to consideration of the requirements fusion must meet to be a viable and attractive energy source. There is concern that when the feasibility of fusion power is established, the federal government will lose interest if a strong argument cannot be made that fusion can be made economical as well.
''We are entering a period of engineering realism,'' says Robert W. Conn, a professor of nuclear engineering at the University of California at Los Angeles. ''We know what we must do. We know we need smaller-size reactors. We know we need simplicity and reliability to make fusion power attractive.''
The question is how to accomplish this. So far, all the engineering work has consisted of paper studies in the attempt to determine the critical issues. Now, there is need for actual engineering studies to determine such issues as radiation damage on various materials, corrosion and thermal shock, how to construct blankets where lithium is converted in tritium, and so on.
BUT, at least in the US, there is little money for such efforts. In October 1980, then-President Carter signed an act outlining an ambitious fusion program. But when President Reagan took office he proclaimed the energy crisis over. As a result, his administration has largely ignored the fusion program, allowing it to continue at the same level but not approving major new efforts.
As a result ''it could easily take 25 years or more to develop fusion into a commercial energy source,'' estimates UCLA's Dr. Conn. ''If we could take the gloves off, in 15 years we could find out whether fusion is real,'' he says. Current discussions about fusion's potentials and drawbacks are based on little more than ''cartoons,'' the engineer objects, adding, ''People are trying to decide at way too early a date.''
Engineering considerations have led the US program to place a major emphasis on a concept called tandem mirrors. The largest machine of this design is the one under construction here at Livermore. It consists of a length of circular magnets, plugged at each end by a pair of large, ''yin yang'' magnets shaped like pinched seams on a baseball.
These are known as mirrors because they reflect the plasma much like an ordinary mirror reflects light. This machine, known as the Mirror Fusion Test Facility, will begin operating in 1987 but is not large enough to achieve ignition.
Beyond this there are a number of advanced concepts under study. ''The most intriguing concepts are the ones we know the least about,'' comments Harold P. Furth, director of Princeton's Plasma Physics Laboratory.
One of these, called the reverse field pinch, has a number of proponents. It creates a much higher density plasma than the tokamak.
As a result, it can be made much smaller than the other approaches. For equivalent sized reactors, the reverse pinch would take up 1/30th of the space and require 1/23rd the material. This could translate into substantially lower cost.
Then there is another, radically different approach. The tokamak and the other concepts already discussed all use magnetic bottles to confine the heated gas. They are called magnetic confinement fusion. These are conceived as continuously burning fusion furnaces. The other approach, called inertial confinement, is more like an internal combustion engine. It is powered by a series of explosions. Only in this case, the detonations are micro-hydrogen-bomb blasts.
IN another building on Livermore's ''back forty,'' the world's most powerful laser, called Nova, is being built for this purpose.
Two-story racks of 30-inch tubes and mirrors route the 10 beams of this monster laser through the portals in a large steel sphere. Within this sphere, these fiery beams of green light converge on a microscopic pellet of deuterium-tritium. This pellet is coated with a layer of material, called rocket fuel, that explodes outward when hit by the laser blast. This applies pressures on the pellet core of 10 to 20 million atmospheres. For an instant, the pellet material becomes the most dense matter on Earth. And this should be enough to detonate a small thermonuclear explosion.
''Inertial confinement fusion can make a high-density reactor of remarkable simplicity,'' argues Michael Monsler, vice-president of KMS Fusion Inc, a company that has been at the forefront of this research.
The Nova machine will push to the very boundary of the conditions needed to trigger a fusion explosion. ''There is some concern that it will achieve ignition, so no one is promising,'' explains Robert L. McCrory, director of the University of Rochester's laser laboratory.
As with magnetic confinement, experts say, inertial confinement's biggest obstacle is economics.
Substantially less engineering work has been done on laser fusion than on magnetic confinement. This is in part because laser fusion's origins are in nuclear weapons research. The basic physics of what happens when a pellet and a nuclear bomb explodes are the same. So US work in this area is funded entirely out of the Department of Energy's weapons program and 90 percent of the work is classified. According to the scientists involved, the Soviets have led the way in declassifying results in this field, not the US.
''Progress in the last three to four years has been spectacular. However, we can't talk about it because it is classified,'' Dr. Emmett complains. ''If we are to continue to make progress, we need a broader range of intellectual talent than we have today.''