High-Magnetic-Field Research and Facilities: [Final Report] (1979)

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.

4 Applications INTRODUCTION Other chapters of this report deal with the opportunities for new scientific research at high magnetic fields and the techniques and facilities necessary for the prosecution of such research. This chapter is concerned with a different aspect-the technological payoff already achieved or expected from the avail- ability of higher fields. Scientific research does not always give birth to new technology; quite often there is considerable delay between the emergence of new knowledge and its exploitation. No such delay occurred in high-magnetic-field research; indeed, the techniques necessary to advance this science are also capable of rapid exploitation in new technology vitally important to future U.S. eco- nomic progress. To illustrate this point, we will mention several national problems for which the technological solution depends to varying degrees, sometimes crucially, on high magnetic fields. That new energy sources are crucial for the future economic health of the United States hardly needs to be noted; almost equally important is the more efficient distribution and use of energy. High magnetic fields can contribute to both of these goals. For example, the most promising long-range solution to the energy supply problem is the achievement of thermonuclear fusion. At present, high magnetic fields provide an attractive method of containing the high-temperature, high-density plasma that is necessary to achieve a fusion reaction. Similarly, the efficiency of electric power generation should be radically improved. One possible way is to use a magnetohydrodynamic (MHD) power generator in which a hot ionized plasma is passed through a magnetic field, generating power directly. The overall efficiency of such an MHD generator increases as the magnetic-field strength is raised. It is also possible to increase the efficiency of electric generators by increasing the strength of the magnetic field in the excitor coils. Later in this chapter we will discuss these and other promising high-field, energy-related technologies in greater detail. 38

Applications 39 National defense technologies are also strongly affected by the availability of higher magnetic fields. The key concept here is that higher field strengths will enable engineers to reduce the size and weight of rotating electrical machines without any loss of power throughput. Essentially, the machine power density is greatly increased, which will almost certainly have a far- reaching influence on the design of propulsion systems for ships. Further, energy-intensive weapons systems will attain lighter, more compact power supplies. What is useful for defense will also apply to many aspects of civilian transportation; radically new systems of propulsion and lift for vehicles will emerge when higher magnetic fields are available. One of the key technological advances that opened some of the possibili- ties we have mentioned was the discovery of high-field superconductors, which occurred during the past 20 years. Because of the relative youth of this technology and its central role in high-field applications, we will discuss briefly how it works and how it fits into the general framework of electrical technology. The rest of the chapter will be devoted to various technologies for which high magnetic fields play an essential part. We have not attempted to catalogue every possible application but have focused on those that seem to us most important and farthest advanced. The following conclusions and recommendations are based on what we believe is required in high-field re- search and development to promote the advance of various technologies and to lead to new technologies at very high fields. CONCLUSIONS We conclude that there are many important, currently developing tech- nologies for which high magnetic fields, in excess of 2 T, are vital to success. In many of these applications, the range from 2 to 10 T has not yet been fully exploited; indeed, in several instances, the achievement of a large working volume is more of a problem at present than higher field strength per se. For this reason, technologists are not pressing strongly for fields above 10 T. Stated another way, the relatively recent access to fields from 2 to 10 T has created so many exciting opportunities for engineers that fields beyond 10 T have not yet been seriously considered by most technologists. Exploitation of still higher fields, beyond 10 T, is likely to yield new technological opportunities. For example, successful mirror machines for fusion would probably require fields above 10T. Increases in the power density of motors and generators could be achieved with higher fields. If fields in the 100-T region were available, new chemical processes would be possible, such as the separation of oxygen from nitrogen in the gaseous state.

40 HIGH-MAGNETIC-FIELD RESEARCH AND FACILITIES Although we cannot point out specific needs for higher fields than are extant, aside from the chemical application mentioned above, the existence of a facility providing such a higher field, whether it be 30 T, 45 T, 60 T, or higher, would be valuable insofar as it would stimulate invention on the part of the applications technologists. It would give them a goal to strive for. In addition, such a facility would afford critical field and current density mea- surements to people working in high-field superconductivity and magnet de- velopment. This would inevitably advance the state of the art and thus add to science and technology. Ideally, one would prefer a steady-state facility, even though quasi-static fields can supply a poor second choice for reliable mea- surements when steady-state fields are unavailable. Our considered opinion is that the health of high-field superconducting magnet technology in the United States would be greatly improved by a research facility providing steady fields up to 45 T. The experience gained in building such a facility, together with the research data that it would yield, would benefit magnet technology across the board, thus having an impact on the many technologies that we will describe in this section. Therefore, in- creased research is required to search for new superconducting materials with upper critical fields in the 45-100 T range, if possible with critical tempera- tures above 20 K. In addition, funds are required specifically to establish the process tech- nology necessary to produce long lengths of practical magnet conductors suitable for superconducting coils in the 20-45 T range. HIGH-MAGNETIC-FIELD TECHNOLOGY The basic building block of a high-field magnet is a multiturn coil of con- ducting material through which an electric current passes. The invention of this device in the nineteenth century led, through the work of Faraday and Maxwell, to the modern electrical industry. Magnet coils are used in gener- ators, motors, transformers, and many other electrical devices. As electric power technology developed, it was found that the electric current used to excite the magnet coils could not be increased indefinitely because of resistive heating in the conductor. However, the magnetic field in the coil could be enhanced by inserting a strong magnetic material such as iron. An optimum mix of copper, iron, and ventilation resulted in machines designed for mag- netic fields up to about 2 T. It is not economic to operate copper-iron systems much above this flux level, which is essentially the saturation field of iron. Of course, steady magnetic fields above 2 T can be readily achieved in the scientific laboratory by forced cooling of copper coils, as Bitter did in his pioneering work. Modern high-field magnets of this type can attain 20 T or

Applications 41 more, but they are not of great use to the electrical engineer because of the enormous power required for operation and the severe cooling problems. In 1911, Kamerlingh Onnes discovered that many metals lose their electri- cal resistance entirely at very low temperatures. For some time it was thought that such superconductors would be used as windings for high-field electro- magnets. Unfortunately, these pure metal superconductors became resistive in quite modest magnetic fields in the vicinity of 0.1 T. In 1961, Kunzler and his associates found a new class of superconductor, now called Type II, which could remain superconducting hi extremely high fields exceeding 20 T. By this discovery, the way was opened to a new era of electrotechnology based on high-field superconducting coils. The progress hi developing superconducting magnets was quite rapid after 1961. Today, 10-T magnets are quite commonplace, and the technology is pushing toward 20 T. As might be expected, these early high-field coils were used primarily hi research. Technological applications generally require large working volumes, as well as high fields. It should be noted that problems of developing coils with larger volumes are at least as formidable as those con- nected with increasing field strength. Most current development work on superconducting magnets is funded as part of major technology programs such as fusion and MHD. The availability of suitable high-field magnets is a crucial element in the overall plan for these new technologies. Support of magnet development in this fashion is highly appropriate, but it can be argued that there is also a tendency in these programs to neglect some of the fundamental research that ought to be pursued in parallel-a point to which we will return. ENERGY TECHNOLOGY The applications of high magnetic fields to improvements in energy tech- nology are diverse and cover a broad range from the national requirement for new energy sources to the technologies of more efficient energy conversion (between heat and electricity, for example), distribution, and utilization. High fields also offer hope of coping more effectively with two serious diffi- culties associated with energy plants, that is, their pollution behavior and reliability. Fusion Reactors One of the most important future energy source options for the United States is the controlled thermonuclear reactor using hydrogen isotopes as a fuel. The basic and unsolved problems of this reactor are achievement of both a suffici-

42 HIGH-MAGNETIC-FIELD RESEARCH AND FACILITIES ently high temperature and adequate density in the fuel plasma to enable a self-sustaining reaction to take place. The plasma cannot be contained in any material container. Possible methods of containment include either a high- magnetic-field "bottle" or the inertia of the plasma itself. Of these two op- tions, magnetic confinement is currently the more advanced in development. Practical fusion reactors with magnetic plasma containment are virtually inconceivable without large superconducting magnets. The Department of Energy is allocating major funds to the toroidal-field Tokamak experiments, with less support to the "magnetic-mirror" experiments. The development of large D-shaped toroidal-field magnets to be used in Tokamak machines has begun; large NbTi solenoidal magnets, baseball magnets, and yin-yang mag- nets are being built for mirror reactors. These magnets are generally in the range of 6-8 T. Also under development for both concepts are Nb3Sn mag- nets intended for fields ranging from 10 to 16 T. Tokamaks are fusion devices in which a solenoidal field, containing a high-temperature plasma, is bent into a torus to eliminate end losses. The toroidal field is provided by D-shaped superconducting magnets with minor diameters typically one fourth of the major diameter. These machines are potentially capable of providing power plants of several gjgawatts at a com- petitive capital cost of about $1000/kW. The smallest economic plant would thus represent a several billion dollar investment. The concept of magnetic mirror containment has recently benefited from two major innovations, each leading to a new range of potential configura- tions and applications. The first of these is the tandem mirror concept, in which a solenoidal field of perhaps 6-8 T contains a cylindrical plasma that is "end plugged" by two pairs of high-field (12-18 T) yin-yang magnets. The relative simplicity of this geometry, particularly when used in a hybrid fis- sion-fusion variant, could offer genuine economic competition to the tor- oidal Tokamak in the multigigawatt range. The second innovation, the recent field-reversed mirror concept, is unique among its competitors, for its ideal individual cell size is of the order of 20 MW, and its geometry, materials, required magnetic fields, and neutral- beam injector would all appear to be of modest cost and to require only modest technology. The achievement of a stable plasma in this configuration is problematical, but experiments within the next two years should provide an answer. The release of kinetically stored flywheel energy into the poloidal fields of Tokamak experiments for ohmic plasma heating is a potential application of advanced rotating-machine technology. Current plans are to use conventional alternators for this purpose; superconductive homopolar machines with high- field excitation offer an attractive alternative.

Applications 43 MHD Power Magnetohydrodynamics (MHD) is a technique to generate electrical power by means of an interaction between a conducting fluid and a magnetic field. The fuel is essentially hot gas derived from coal, oil, or a nuclear reactor. The motivation for the development of MHD generators for use in central station power generation is the potential for a significant improvement in plant ther- mal efficiency as a result of the much higher temperatures that are used in MHD generators. The power density that can be attained in an MHD generator is pro- portional to the electrical conductivity of the fluid and to the square of the flux density of the magnetic field. Because the conductivities of the fluids that are available for use in practical MHD generators are quite low, seeding chemicals are injected into the fluid flow to increase these conductivities. However, even with the use of seed, the conductivities of these fluids remain several orders of magnitude below that of copper. As a consequence, high magnetic fields are required to maintain the physical dimensions of these generators within practical limits. When coal is burned to provide combustion gas as the fluid, the seeding materials have the added feature of removing virtually all the sulfur oxides from this gas, thus substantially reducing efflu- ent pollutants. The power required to generate magnetic fields in excess of 2 T over large volumes becomes prohibitively high if these fields are generated by making use of conventional conductors for their windings. Consequently, making use of conventional magnets would reduce the efficiency of MHD generators to a level that would render them unattractive for use in central station power generation. On the other hand, electromagnets making use of super- conductors for their windings can readily generate magnetic fields in the range of 5-7 T and require hardly any electrical power to maintain these fields. High magnetic fields generated by means of superconducting windings are, therefore, essential to the successful development of efficient MHD generators for central station use. Because 40 percent or more of the U.S. power plants are expected to be coal-fired by the year 2000, the MHD program emphasizes the use of coal as a fuel. The United States currently hopes to have an operational pilot plant making use of MHD by 1985 and a base-load commercial demonstration plant by 1995. To date, the largest superconducting magnet built for use in an MHD generator weighs 40 tons and develops a magnetic flux density of 5 T. This magnet was constructed by the Argonne National Laboratory for test pur- poses in a joint U.S./U.S.S.R. (U-25) experimental program to develop MHD power generators. Design studies have been completed on the super-

44 HIGH-MAGNETIC-FIELD RESEARCH AND FACILITIES conducting magnets required for full-scale (600-MW) central-station-type MHD generators. These studies reveal that such magnets will weigh approxi- mately 2200 metric tons and cost about $60 million. Plans are to work at about 6 T in these systems, but higher fields would be desirable for higher efficiency if they could be attained reliably and economically. The MHD combustion studies have indicated that the plant and process would be simplified if one could work with a combustion airstream enriched with oxygen. A possible approach to achieving such enrichment is to use the strong paramagnetism of the oxygen molecule, which enables it to be separ- ated from nitrogen by a high magnetic field (perhaps the main MHD magnet field). This process is under study in several laboratories; preliminary work suggests that reasonable efficiencies may require fields in excess of 100 T. It appears possible that if fields of this magnitude were readily available in the engineering sense, a variety of new chemical separation processes of this type would become economically feasible. Electrical Machines As we have noted, modern electric machine technology is founded on cop- per-iron electromagnets operating up to about 2 T. Most machines are air cooled or, in the case of central station generators, cooled by hydrogen gas or water. Because of the convenience of voltage transformation, the bulk of the present power systems are based on ac machines, with dc confined to special applications such as railroad traction. With the advent of high-field supermagnets in the 1960's, a search began for the optimum areas of machine technology in which superconductors could replace copper systems. The pattern that has emerged so far is that the supermagnets are not suitable at present for direct substitution in ac windings because of excessive losses in the superconductors that are currently available. However, there are specific devices such as dc homopolar machines or ac synchronous generators in which use of a high field in part of the machine would be highly advantageous. The dc homopolar machine was invented by Faraday. Its principal feature is a disk or drum rotating in a magnetic field applied normal to the plane of the disk, causing current to flow from the axis to the periphery of the disk. Several high-field superconducting experimental models of this machine were constructed in the past decade, but despite savings in size and weight, it has not been adopted for widespread industrial use. The most promising applica- tion at present seems to be ship propulsion, which we discuss subsequently. A different situation exists for the central-power-station, turbine-driven ac synchronous generator, in which a large rotating dc field winding generates ac power in a stationary external armature. The power dissipation in the rotor is

Applications 45 almost half the power loss of the entire machine and can be essentially eliminated by substituting a high-field superconducting winding for the pres- ent system of copper coils mounted in a slotted ferromagnetic steel rotor. Several medium-size experimental prototype superconducting machines were constructed in the early 1970's, specifically at MIT (3 MVA) and Westing- house (5 MVA). The Electric Power Research Institute (EPRI) has recently supported design studies for a 300-MW machine and will probably contract for the development of such a machine in the near future. The magnet will probably be wound from a niobium-titanium superconductor and will oper- ate at about 5-T maximum field. The economic benefits of using superconductive high magnetic fields in central station generators are quite impressive. Not only are the generator losses reduced by 50 percent, but the higher magnetic field allows the frame size to be reduced, lowering the overall cost of the machine appreciably. A further possible benefit is the addition of a high-voltage stator, which will enable direct generation of transmission line voltage, thus eliminating a large step-up transformer. Benefits could exceed $25 per kilowatt for a large power plant. If such machines capture a major fraction of the U.S. market in the next two decades, utility revenue needs through the year 2000 should be reduced by about $2 billion. The design philosophy of these early machines is heavily influenced by the extreme need for reliability in power station generators. The superconducting magnet design tends, therefore, to follow a well-tried engineering path, with a field strength (5 T) well short of what is known to be achievable using more advanced superconductors. Wider use of high fields in machines and transformers might develop if the ac losses of high-field superconductors could be reduced or eliminated. This research has not received a great deal of attention and would seem to be a fruitful direction for future effort. Energy Storage A major problem of electric utility systems is that a good fraction of their generating plant is idle much of the day. This underused equipment is called on for only short periods to supply peak loads. Electricity costs could be reduced if electrical energy could be stored in low load periods and released at peak times, so that the generating plant could be run continuously. Electri- cal energy can be stored in batteries, capacitors, and high-field super- conducting magnets. Recently, at Los Alamos and the University of Wiscon- sin, serious studies of major energy-storage systems using superconductivity have taken place.

46 HIGH-MAGNETIC-FIELD RESEARCH AND FACILITIES These energy-storage projects are at too early a stage of development to allow firm statements about their economic feasibility. Preliminary con- clusions suggest that the load-leveling requirements of the United States, which are principally diurnal, could be satisfied by Inductor Converter units covering about 10 percent of the peak electrical load. These magnets would probably be extremely large-volume units constructed from cryostable Nb-Ti conductors, with operating fields below 5 T. The future needs in this instance are for higher-current-density, lower-cost superconductors, as opposed to higher-field systems. Energy Utilization The application of high magnetic fields to energy utilization has received less attention, probably because most end-use devices are small and widely distri- buted. An exception is public transportation systems. Some radically new concepts have emerged in the past two decades based on the use of high magnetic fields to suspend vehicles magnetically in place of conventional spring and wheel systems. Magnetic suspension for vehicles is an old idea, dating from Bachelet's work early in the 1900's. The basic principle is that a repulsive force is produced between a magnet (on the vehicle) and eddy currents generated in a conducting track over which the vehicle moves. The concept does not seem to be practical for low-field magnets, below 2 T; however, it becomes economi- cally more attractive when high-field superconducting magnets are employed, as Powell and Danby pointed out in 1966. Magnetic suspension has several advantages over more conventional tracked vehicles. With conventional tracked vehicles, major problems arise at speeds over 300 km/h. Track alignment becomes critical and hard to main- tain; excessive wear occurs between rail and wheels and also at sliding electri- cal contacts for electric propulsion. Separation between train and guide rails is achievable with magnetic levitation, and magnetic propulsion can be ob- tained by a linear-induction or linear-synchronous-motor arrangement in the guide rail. Small-scale models of magnetically levitated vehicles have been tested in the United States, Germany, and Japan. Larger systems are under development, particularly by the Japanese National Railways, which is now the world leader in this technology. The National Aeronautics and Space Administration (NASA) has recently been exploring an interesting application of high magnetic fields to heat pumps, another field of energy use. This application is based on the principle of adiabatic demagnetization of paramagnetic materials, which has long been known to low-temperature physicists. The same principle can be applied at

Applications 47 room temperature using gadolinium, which has a Curie temperature of 20°C. By use of a recuperator and counterflow heat exchangers in conjunction with a high, varying magnetic field, a Stirling cycle heat pump ensues; in the high-field limit this concept has Carnot efficiency. The basic principles of this device were demonstrated at NASA; current funding is provided by EPRI. The NASA demonstration showed that a re- cuperator could develop a temperature differential exceeding 80 K in a 7-T field; the EPRI experiment will demonstrate a prototype heat pump using this principle. The efficiency should be several times higher than that of conventional heat pumps and refrigerators. MAGNETIC SEPARATION Low-field electromagnets have been used for many years as a general tool for iron scrap separation and ore beneficiation. The promise of higher fields has recently stimulated great interest in this subject, which is focused on mag- netic separation of much more sophisticated mixtures than ever before. A promising energy application is the removal of pollutants from the cooling system of power plants and also from fossil fuels prior to combustion. The technology involved is High Gradient Magnetic Separation (HGMS). The principle of HGMS is the interposition into the working field volume of a porous matrix of magnetizable material so that many high-gradient interfaces are present. Stacked iron balls and ferromagnetic steel wool are examples. The imposed field should be sufficient to magnetically saturate the matrix; for most separations the efficiency continues to increase with increasing field by raising the magnetization of the paramagnetic particles being collected. The use of magnetic filters is growing most rapidly in nuclear-power- station technology. The corrosion products are fine-magnetic-particle oxides and hydroxides of construction metal elements such as iron, manganese, and nickel. Radioactive species of various origins also appear in these products. Filtration takes place at ~300°C, and flow rates range from 10 tons/h in test units up to projected rates of 3600 tons/h. Development of this application started in 1970, with matrices of steel balls, which provided mediocre magnetic gradients. Recently, there have been tests with the superior steel wool matrices. Current field levels are modest, 0.5 T, but over a large volume of 2-m diameter by about 0.5 m high. Heit- mann forecasts that further development will use superconductive coils for higher fields and ultimate separation of weaker paramagnetic materials. In contrast, Oberteuffer believes that industry would be slow to adopt super- conductive technology and that the cheap power available at power stations reduces the economic stimulus for conversion to superconductivity.

48 HIGH-MAGNETIC-FIELD RESEARCH AND FACILITIES Another possible application of HGMS is to the desulfurization of coal. Between one third and two thirds of the sulfur in eastern and central U.S. coals is bound as discrete particles of inorganic compounds of sulfur and iron, principally iron pyrites, which are liberated when coal is pulverized before combustion. The HGMS approach to coal cleanup is one of a number cur- rently under consideration that have varying costs, efficiencies, and problems. Its low add-on cost, estimated at roughly $1 to $3/ton of coal, makes it attractive. By contrast, the well-known flue gas desulfurization adds about $10 to $15/ton to coal costs. A variety of chemical coal cleanup methods are also under consideration, with estimated costs of from $3 to $10/ton. A promising and related example of HGMS is the removal of inorganic mineral particulates from liquified coal. Although such coal conversion pro- cesses appear to be expensive, that is, $10 to $30/ton, they are receiving much attention and major funding from the Department of Energy, EPRI, and oil and utility companies because of the useful form of the product, for example, as liquid or easily transported solid. As with the magnetic desulfuri- zation of raw coal mentioned in the preceding paragraph, the particulate cleanup of liquified coal is attractive from an economic viewpoint. Oder has estimated a cost of about $ 1 or $2/ton of coal, a factor of 4 cheaper than the current methods of high-temperature filtration. In addition, the chemical process of hydrogenative liquefaction renders the iron-sulfur compound more strongly magnetic, that is, conversion of pyrite to pyrrhotite. Other applications for HGMS include mineral processing and ore benefici- ation, which has a long history if one includes poorer-quality separators (Edison and others before him). The improvements brought about by HGMS permit beneficiation of more weakly magnetic ores such as oxidized taconite iron ore and the separation of paramagnetic tourmaline from the tin mineral cassiterite, Sn02. Waste-water cleanup is also feasible either by the direct removal of suspended magnetic solids, as from steel mills (see also power station water discussed previously) or by the coagulation of contaminants onto magnetic "seed" particles. This process has been demonstrated for oils, phenols, coliform, PCB's, and other water impurities. There are also studies of cleanup of magnetic particulates from stack gas streams, for example, from steel-making furnaces. Another novel and interesting technique, with an open but lower gradient coupled with a long-interaction path, has been developed in the United Kingdom. It uses a superconductive quadrupole coil arrange- ment (H £ 6 T) with a spiral flow channel. This approach is especially attrac- tive when the relative fraction of magnetic material is no longer small. All of these applications can be classified as magnetic separation of the first kind, relying on the inherent magnetic susceptibility of the material to be separated. Magnetic separation of the second kind occurs when the medium of separation rather than the separated particles is made magnetiz-

Applications 49 able, for example, a colloidal solution of a ferromagnetic or ferromagnetic substance, or an aqueous solution (or melt) of a strongly paramagnetic salt. The apparent density, hence the effective buoyancy force, can be adjusted by the applied magnetic-field gradient. Such techniques are under study and development with iron electromagnet systems for reclaiming various non- ferrous metals from scrap and for ore beneficiation. Another variation, eddy- current separation, has been developed using permanent magnet arrays. It is finding practical application in removal of nonferrous metals from solid waste. DEFENSE TECHNOLOGY Ship Propulsion The British Navy substantially advanced superconducting machine technology by providing funds for the development of a 50-HP superconducting solid- brush-switched homopolar dc motor that operated successfully in 1965. The design was expanded to a 3250-HP pump motor that was demonstrated in 1971, and to a 1-MW land-based model of a ship power transmission sys- tem in the mid-1970's. These machines contained no iron and emphasized low weight; small machine diameter, high efficiency, and low helium con- sumption were of lower priority. The U.S. Navy superconductive electric propulsion program started in the late 1960's and emphasized high efficiency, which implied low-helium- consumption, small-diameter motors to permit fitting them in low-drag na- celles, and adaptability to driving high-efficiency, contrarotating propellers. The resulting homopolar acyclic dc machines have liquid-metal current collec- tors and are compact, rugged, and simple, with liquid helium consumption of the order of 5 liters/h and efficiencies of 98-99 percent. A 400-HP motor of this type has served as an experimental facility for five years; several 3000-HP propulsion systems are under construction and will probably be operated in the laboratory and at sea during the next two years. Systems of 40,000 HP are planned for the early 1980's. The synergistic benefits resulting from the flexibility of arrangement and operation permitted by these efficient, compact systems could reduce fuel consumption of naval monohull combatants by as much as one half for the same payload, performance, and range. (The cost of fueling U.S. surface ships is $2 billion per year.) Equally important, the lightness and compactness of superconducting systems could be essential to the economic feasibility of long-range, high-performance ships. The development of practical advanced superconductors such as NbjSn or V3Ga could further benefit naval machin-

50 HIGH-MAGNETIC-FIELD RESEARCH AND FACILITIES ery by providing the ultimate propulsion system for high-performance ships. A subtler, less-quantifiable advantage is the greater operating margin (equiva- lent to a "factor of safety" in other fields), which would be highly regarded by any ship's commanding officer. Part of the superconductive machinery program is the development of compact, shock-resistant helium refrigerators and compressors of 10 liters/h capacity. High-power-density oil-cooled rectified alternators are also being built as alternatives to acyclic generators. The development of other advanced normal-temperature, liquid-cooled machines has been stimulated by this pro- gram. Airborne Electric Power Future combat aircraft may require electrical power levels of tens of mega- watts for periods of several seconds. To produce such power, chemical energy stored in propellants may be converted into electrical power by either turbo- alternators or MHD devices. Obviously there is a premium on the achievement of compact, light systems for this application. The U.S. Air Force has funded programs on high-speed (6000-12,000 rpm) superconductive alternators since the early 1970's. Severe conditions are imposed: very rapid speed changes (several thousand rpm per second), high radial accelerations of liquid helium in the rotor (above 10,000 g), and high coil-charging rates (several T per second). Under these severe conditions, NbTi superconductors are marginal; the development of multifilamentary Nb3Sn wire with high-temperature capability thus received its first major impetus from a U.S. Air Force development program. Permanent-magnet alternators are strong potential competitors in this field, as are plasma MHD systems using superconducting magnets. These two systems are under development by the Air Force. High-Frequency Radiation The ability to generate appreciable power levels of high-frequency radiation in the broad region between radio frequencies and the optical band has opened radically new defense technologies, of which radar is probably the most prominent. One of the key devices that made microwave radar possible is the cavity magnetron, in which electrons are caused to move in circular orbits by an applied magnetic field, usually provided by a large permanent magnet. The possibility of creating other novel electron-radiation devices by using much stronger magnetic fields has intrigued inventors for many years. One

Applications 51 TABLE 2 Emission Wavelength versus Magnetic Field and Cyclotron Harmonic Number Emission Magnetic Field Wavelength 10T 30 T 75 T 120T 1000T co = coc, X = 1072 Mm 357 Mm 143 Mm 89 Mm 10.72 Mm CO = 2U)C, X = 536 Mm 179 Mm 71 Mm 45 Mm 5. 36 Mm to = 3coc, X = 357 Mm 119 Mm 48 Mm 30 Mm 3. 57 Mm co = 4coc, X = 268 Mm 89 Mm 36 Mm 22 Mm 2.68 Mm such device is an electron resonance maser; this is an electron-beam tube operating in a uniform magnetic field and emitting coherent radiation at the electron cyclotron frequency, coc or eB/mec, and its harmonics. The electron mass, me, must be treated relativistically for an accurate frequency calcula- tion. In practice, most electron cyclotron masers have operated in the weakly relativistic regime, with energies of 20-80 keV. A particularly efficient high- power version of the electron cyclotron resonance maser, called the gyrotron, has recently been invented in the Soviet Union and operated at wavelengths as short as 2 mm in the fundamental (coc) operation and 0.92 mm in the first harmonic (2coc) operation. Because the emission frequency depends linearly on B, it is obvious that high emission frequencies (that is, short wavelengths) can be achieved at high magnetic fields. Typical numbers appear in Table 2. Short wavelengths might be achieved by either increasing B or harmonic number n or both. However, efficiency drops off rapidly with n, so that if efficiency greater than 10 percent is required, operation at coc (« = 1) will probably be necessary; therefore, high magnetic fields will be required for high frequencies. The gyrotron is the most promising device for extending high-power microwave capabilities to higher frequencies. Aside from the defense and communication uses, another potential application includes heating magnetic- ally confined plasmas to very high temperatures to initiate thermonuclear reactions. A fusion power plant might require 100 MW or more of microwave heating power at a frequency of the order of 200 GHz, which might be supplied by a bank of 100-kW gyrotrons. OTHER APPLICATIONS There are several applications of high magnetic fields other than for energy or defense. These include the use of fields in certain specialized materials tech-

52 HIGH-MAGNETIC-FIELD RESEARCH AND FACILITIES nologies and the development of new particle accelerators, which, although they are primarily research devices, could yield exciting future technological payoffs. Materials Technology High-field magnets were important in the recent discovery of a new class of permanent magnet materials based on intermetallic compounds of rare earths and cobalt. In 1966, the magnetocrystalline anisotropy of a single crystal of YCos was investigated using a 5-T superconducting solenoid. The conclusions of this work pointed to prospects for new permanent magnet materials and stimulated a major effort in materials research and metallurgical processing. The high coercivities, up to 3 T, of various members of the rare-earth cobalt family require high fields for the initial magnetic alignment process, as well as in characterization studies. The figure of merit of this permanent magnet system is roughly double that of its predecessors. This increase permits great changes in magnetic cir- cuit design, together with considerable savings in power, weight, and space. An example of these savings is the development of a new 8-inch-diameter battery-powered torpedo motor of 140 HP to replace a previous version of at least 12-inch diameter. The detection of magnetic impurities in solids can be done with increasing sensitivity as the applied magnetic field is increased. So far, this technique has been of value chiefly as a means of studying weak magnetic materials in the laboratory; however, for the first time, the General Electric Company re- cently applied it in a manufacturing plant. A 7.5-T superconducting magnet system is employed as part of a system to monitor planned variations of magnetic properties in nuclear fuel rods. This type of quality-control function using high magnetic fields will probably become more widespread in the future. The superconductors that are used in the construction of high-field mag- nets must themselves be evaluated in the high-field environment. Lest this be regarded as merely a "bootstrap" operation, it should be noted that the development of these materials over the past 20 years depended heavily on the availability of water-cooled copper solenoids and pulsed-field magnets. Measurements up to 60 T are necessary, and these can only be made with pulsed fields at present. High-Energy Particle Accelerators The application of high magnetic fields to beam bending in high-energy particle accelerators might seem to be of primary interest for research in particle physics. Although this is true, there has also been synergism between

Applications 53 the development of high-field devices for nuclear research and the general development of large high-field magnets; indeed, in the past two decades progress in magnet engineering has come about more from nuclear interest than from any other source. Most of the accelerators have been based on copper-iron technology and limited to working fields of around 2 T. The advantages of higher fields are obvious; that is, the total size of machine can be reduced while maintaining the desired energy level. Alternatively, larger particle energies can be achieved with a given size of machine. A further advantage for large machines is that operating power costs can be reduced by using superconducting magnet windings. Accelerator engineers have been reluctant to convert to major super- conducting systems until these were shown to be reliable and predictable from an engineering viewpoint. Therefore, they have taken tentative steps in the new direction by building auxiliary devices such as bubble-chamber mag- nets and individual sections of accelerators. This stage now seems to have passed, with designers moving ahead to full-size machines. For example, a new energy doubler or Tevatron is being constructed at Batavia, Illinois, in the same tunnel used for the present machine. By doubling the deflecting field to about 4 T, the bending radius is maintained for the increased energy. Using Nb-Ti magnets, the energy can be doubled for approximately $30 million, which is about one tenth of the original accelerator cost. The next large-scale application of this new high-field technology is Isabelle at Brook- haven National Laboratory. This high-energy accelerator will operate at 5 T and will generate colliding beams at 400 GeV each. The cost of $275 million is again small when compared with that required by conventional high-field technology. Future applications of accelerator development offer a number of inter- esting technological possibilities for the future, particularly if the size and operating cost can be greatly reduced by such methods as those mentioned in the preceding paragraph. Accelerators (and storage rings) are already used as sources of continuously tunable high-frequency (ultraviolet and x-ray) synchrotron radiation for basic and applied research. The insertion into the electron beam of high-magnetic-field superconducting "wiggler" magnets (so- called because the fields wiggle the electron beam, causing it to emit further radiation) enables users to tailor the spectrum of radiation to specific require- ments. Much work on the development of these magnets at the highest possi- ble fields is required. High-energy heavy ions have also been proposed as a trigger for controlled thermonuclear fusion. The ions would deposit their energy into pellets of thermonuclear fuel, heating them to fusion ignition temperature. The high energies (~ 200 GeV) and high currents required would lead to an enormous

54 HIGH-MAGNETIC-FIELD RESEARCH AND FACILITIES accelerator unless high fields were used to bend the beams to within a man- ageable radius at a manageable cost. Finally, high-energy, high-current proton accelerators are being considered as a means of burning used fuel from nuclear reactors, breeding new fissile materials, and producing energy by inducing nuclear reactions on impact of the protons on the radioactive material.