Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
PROSPECTS FOR MAGNETIC CONFINEMENT Some 25 years have elapsed since the U.S., Great Britain, and the Soviet Union each embarked on a serious program to harness fusion for civilian purposes. Many other nations joined in the effort after most elements of the program became declassified in l958. Much progress has been made scientifically and a great deal learned about what has emerged as a new branch of physics, namely plasma physics. Nevertheless, as we shall indicate, it is still premature to judge the practical merits of fusion. Three stages of accomplishment are necessary before a final deter- mination can be reached: l. Scientific Feasibility - wherein reactor-grade plasmas are attained; scaling laws are well understood; and break-even criteria, corresponding to nT = l0 trillion to l00 trillion at plasma temperatures of 5 thousand electron volts (KeV), or 60 million degrees Celsius, and above, are demonstrated. We note that nT value? at the lower end of this range (i.e. nT = 20 trillion) have been achieved in ALCATOR at l KeV ion temperature. 2. Engineering Feasibility - wherein it is demonstrated that a suitably designed power-producing reactor can be constructed and successfully operated, with due regard to safety and environmental impact. 3. Commercial Feasibility - wherein it is demonstrated that reactors of proper design will have all the features necessary to make them potentially competitive, in economic terms, with alternative commercial energy sources. SCIENTIFIC FEASIBILITY In spite of the very considerable progress made on what has proved to be a most difficult problem, fusion research has yet to complete the first phase, that of demonstrating scientific feasibility. There are strong indications, however, that this stage will be arrived at within the next several years, and that most likely it will be the Tokamak series of experiments in which initial scientific feasibility will be shown. The understanding of the behavior of plasmas in Tokamaks is well
l0 advanced, but experiments have been restricted to density and tempera- ture ranges that are short of simulating reactor conditions. The partic- ular forms of microturbulence which limit plasma lifetimes and density are strong functions of temperature and configurational detail. Con- sequently, it is particularly important to extend the experimental studies of scaling laws to the temperature regimes more representative of igni- tion conditions, and it is expected that this will be accomplished in the l977 to l978 time frame when the energetic neutral beam injectors come on line in the Princeton Large Torus (PLT) at Princeton Plasma Physics Laboratory (PPPL) and ORMAK Upgrade at Oak Ridge National Labor- atory (ORNL) experiments. In the event that impurity control proves to be a determining factor in setting confinement times and nT values, the above experiments may be of more limited value. Then, we shall have to await results from the Poloidal Divertor Experiment (PDX) at PPPL and from the Impurities Studies Experiments (ISX) at ORNL in the l978- l980 time frame. For purposes of economic reactor design, implying reasonable power densities, it is important that means be found to increase the 8 value in Tokamaks. The noncircular cross-section experiments in Doublet III at General Atomic Company (GA) and PDX at PPPL, as well as the flux- conserving mode of ORMAK Upgrade at ORNL, should shed some light on this subject during the l978-l980 time frame. Finally, the Tokamak Fusion Test Reactor (TFTR) at Princeton, expected to be operating by l980-l981, should be able to demonstrate plasma behavior under nearly fully simulated reactor conditions, although perhaps short of a true ignition mode.* Thus, it is quite likely that the scientific basis for a fusion reactor design based on the Tokamak principle will be in hand within the next 5 to 6 years, though it may still be short of the opti- mum design. Within this time frame, it may be expected that further confirmatory results will be obtained from similar large Tokamak exper- iments being planned in Europe, the Soviet Union, and Japan. It may be of value when trying to form some outside judgment on the progress and rate of progress in fusion reactor development to note that the PLT, ORMAK Upgrade, PDX, Doublet III, and TFTR devices mentioned above are physics experiments. They are, however, somewhat unusual when compared to many scientific experiments outside the field of plasma physics. Each, with the exception of TFTR, is in the cost range of $l0-$30 million, takes about 3 years to design and construct, and has a useful experimental life of perhaps 3 to 4 years. The operating costs of these experiments are in the range of several million dollars per year. TFTR is unique because of its large size and because it is de- signed to handle tritium. Its cost is one order of magnitude higher than that of the earlier and smaller devices. A small fraction of the total plant cost may be ascribed to the fact that TFTR will be able to handle tritium, in addition to hydrogen. In other words, significant *As originally conceived, Doublet III has the design potential for reaching the ignition criterion; however, present plans do not call for fully implementing this potential
ll advances in aspects of scientific knowledge may occur at 3-5 year inter- vals at costs in the range of l0 million to l00 million per experiment. Figure l represents some of the past and future milestones on the road to achieving scientific feasibility. Results obtained in the magnetic mirror program from the 2X experi- ments at Livermore during l975-l976 have been most encouraging. There seems to be convincing evidence that the observed scaling follows theo- retical predictions and that confinement time increases with the three- halves power of the temperature. Theory also predicts that better control of instabilities will result from increased plasma radius. There seems to be a sufficient basis for constructing a larger device, designated as MX by the Livermore group, in order to extend the scaling studies to regimes where a higher value of nx may be obtained. In order to make the magnetic mirror concept appear attractive for reactor appli- cations, however, it will probably be necessary to find some way of decreasing end losses appreciably. Unless this can be done, the ratio of circulating power to net output power of a mirror reactor would be high and unattractive. Several proposals to reduce the losses inherent to open-ended magnetic mirror configurations are receiving consideration. For example, the tandem mirror scheme for reversing the ambipolar poten- tial (proposed at Lawrence Livermore Laboratory (LLL) and Novosibirsk), the Elmo Bumpy Torus (operating at ORNL), or the field reversal concept under investigation (at Cornell University and LLL), represent, in principle, solutions of the end-loss problem. We note again that while scientific feasibility may be demonstrated within the next several years in one or more devices, any reactor based on these results may turn out to be far from optimum in terms of the performance attainable eventually in practical fusion reactors. However, demonstration of scientific feasibility, besides providing a test bed for the performance of relevant physics experiments, should provide a realistic basis for further optimization of designs leading to economi- cally attractive reactors. ENGINEERING FEASIBILITY Problems relating to the engineering feasibility of fusion reactors operating on the D-T fuel cycle are beginning to be addressed seriously. In this portion we merely summarize certain technological issues which will be discussed more fully in the chapter of technological considera- tions. A number of generic problems exist and are being studied in D-T fusion systems, independent of specific reactor design. These include the behavior of structural material in the intense radiation environment characteristic of a D-T fusion-reactor plasma; superconducting coil de- sign on a large scale, tritium handling, and blanket design. Some amount of information useful to the fusion program is also becoming available from the materials studies that are part of the fast fission breeder program. Both Tokamaks and mirrors will require energetic neutral beam sources or radio-frequency sources, or both, for heating. Neutral beams are the current favorites, and it will become necessary to develop sources capable of operating in the l0-l00 MW range at injection energies
l2 FUTURE Goal for Power Reactors -1 1976 ORMAI01976 ⢠*"*^. THETA PINCHES Ignition 10 = Gain 100keV TEMPERATURE FIGURE 1 The path to scientific feasibility.
l3 of l00-500 keV in pulsed and continuous wave (cw) modes. In addition, certain device-specific problems will have to be solved. For example, Tokamak reactors are conceived today to operate in a cyclic mode, with burn (reaction) times exceeding energy confinement times by large factors. The problem of fueling Tokamaks during the extended burn time remains to be solved. If one were to embark on other than the D-T fuel cycle, recognizing the greater amount of physics uncertainty this would entail, many of the engineering problems listed above that result from tritium handling or neutron activation and damage could be eased or eliminated. At the same time, another set of engineering problems would be introduced, possibly of less severity.
