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.
METHODS OF APPROACH Two main lines of approach towards developing a practical fusion reactor for civilian applications have evolved in the course of the past quarter of a century: the magnetic confinement approach and the inertial con- finement approach MAGNETIC CONFINEMENT In a macroscopic sense, plasmas behave as if they were diamagnetic. Con- sequently, a properly designed magnetic field configuration will produce stresses to counterbalance the pressure of the plasma. In this manner the magnetic field can act as a thermal insulator and thus confine the plasma materially to a specific volume in space. Such configurations are termed magnetic bottles or containers. In principle, radiation pressure could achieve the same result, but in practice one finds mag- netic fields that are either static or slowly varying in time to be the most effective. One measure of the economy by which magnetic fields con- fine a plasma is the parameter beta (3), defined as the ratio of plasma pressure to the magnetic pressure that would exist locally if the plasma were absent. By this simple measure, a beta of unity is the maximum achievable value. Confinement schemes tend to divide into low $ (<0.l) or high 3 (>0.5) categories. Since for a given magnetic field strength and plasma temperature the power density in a reactor scales as B2, the value of 3 achievable is an important design factor in magnetic confine- ment. The usual figure of merit in the magnetic confinement approach, however, is the product of plasma density (n) and energy confinement time (T). When n is measured in particles per cubic centimeter and T is measured in seconds, it can be shown that at fusion reactor temper- atures (on the order of l00 million degrees Celsius) most magnetic con- finement schemes require nT values in the range of l0 trillion to l00 trillion to reach the "energy break-even" conditions, where the energy released through fusion reactions just equals the energy invested in the plasma. An operating fusion reactor would have to achieve nT values somewhat higher than the break-even condition (or Lawson criterion) and in addition, it would have to maintain temperatures high enough to burn a significant fraction of its fusion fuel. Certain early studies indicate
that for economic net power production one is interested in operating the reactor in an "ignited" mode. Under these conditions the 3.5 MeV alpha particles (helium nuclei) produced in the fusion reaction are con- fined in the plasma long enough to heat the intially cold fuel. The nuclear process which then takes place is loosely analogous to chemical combustion. The ignition temperature for the D-T reaction is nominally 60 million degrees Celsius. It is also possible to "drive" a fusion reactor by injecting high-energy particles from an external source into the plasma. Such schemes may in fact be advantageous when one is con- cerned with optimum means for neutron production, rather than net power production. A magnetic field confines charged particles to a small orbit in the plane normal to the field. In order to prevent them from leaking out along the field lines, one must either (a) use a toroidal geometry in which the field lines form closed-flux surfaces, or (b) rely on magnetic mirrors or other end-stoppering effects to reflect the particles. Thus, we can subdivide magnetic confinement schemes into closed and upon con- figurations, of which the Tokamak and mirror, discussed below, are representative. The principal scientific limitation on fusion prospects is the multi- tude of plasma instabilities which may occur. These produce collective currents and charges and fluctuating fields which may destroy the con- finement properties of the system. A vast amount of work has been done and considerable understanding gained on these matters over the years. The most dangerous and rapid instabilities, characteristic of early experiments, are now routinely avoided while some, leading to microtur- bulence, still remain and degrade confinement by a considerable factor. Nonetheless, it is now clear that a plasma which is magnetically confined in a closed system of sufficient size would exceed break-even, although we cannot state precisely at this time what the minimum size must be. Open systems have a smaller safety margin for confinement, but offer the possible advantages of higher B and simpler engineering. Many different magnetic confinement schemes have been explored in the past. The one showing the greatest scientific promise today is a concept pioneered by workers in the Soviet Union, the Tokamak. Tokamak is a toroidal device in which a combination of externally applied toroidal magnetic fields and poloidal magnetic fields induced by toroidal currents flowing in the plasma create the desired magnetic configurations. Another promising confinement concept is the magnetic mirror, in its simplest configuration an open-ended device in which the confining magnetic fields are generated externally by suitably shaped coils. Both the Tokamak and mirror are essentially long-confinement-time concepts, where in an operating reactor one would expect to achieve energy confinement times on the order of several seconds. The Tokamak would probably operate in a cyclic mode with a high duty cycle, while the mirror could operate in a stead state. As now conceived, the Tokamak will operate at lower (3 values than mirrors, which have operated in the high 6 regime. One of the most important features of fusion physics is the enormous range of possible confinement variants. Several of the alternative magnetic confinement schemes (other than the Tokamak or mirror) have
not been studied extensively enough to determine their prospective merits. Some, on the other hand, have encountered unresolved difficulties. With more ingenuity, experience, and understanding, it is conceivable that other fusion systems may evolve with better stability, confinement, and power density properties and lower plant capital costs than those under current investigation. Thus, it is very difficult to assess the ulti- mate character of magnetic confinement systems at this time. INERTIAL CONFINEMENT The range of plasma densities encountered in magnetic confinement schemes may vary from l00 billion cubic centimeters to l0 quadrillion per cubic centimeter; consequently, the critical confinement times for break-even may vary from several seconds to perhaps one-thousandth of a second. Inertial confinement, however, conceives of working at the extreme high density and very short time-scale end of the nT range. Fuel, arranged in the form of a pellet, is rapidly compressed to densities of three to four orders of magnitude above liquid density values (n ~ l0 septillion to l00 septillion per cubic centimeter) in this approach. During the compression stage, the fuel is adiabatically heated to ignition temper- atures at the center of the fuel assembly. Thereafter, a fusion burn front propagates outward. The fuel continues to burn until the highly compressed plasma disassemblies, on time scales (R/v) comparable to the radius of the pellet at ignition (R) divided by the speed of sound (v) in the material. From the earlier nT criterion, one may show that an equivalent figure of merit for inertial confinement is PR, where p is the mass density and R is the pellet radius, both measured at maximum compression. For D-T the pR value corresponding to break-even is on the order of l gram per square centimeter. In order to achieve rapid compression, heating, and subsequent ther- monuclear burn, it is necessary to find some means for delivering a short burst of energy to the pellet. This will cause the surface layer of the pellet to ablate, producing compression by the forces of reaction. The first energy source tried in an attempt to initiate these micro- explosions was the laser; consequently, the inertial confinement approach is often referred to as laser fusion. More recently, energetic electron beams have also been used to drive the implosion; and ion beams, produced by technology borrowed from the electron beam work or by more conventional accelerators, have also been proposed as drivers. The driver energy required to reach break-even conditions depends in detail on the mechanism of energy deposition within the pellet and the efficiency with which this couples to the hydrodynamic phase of compression. Estimates ranging from hundreds of kilojoules to several megajoules have been obtained from computer codes which try to model the various physical processes taking place. The energy will have to be delivered over short enough times to reach power levels on the order of a hundred terawatts (>l00 trillion watts) and focused to flux levels of l0 quadrillion watts per square centimeter.
8 An operating fusion reactor based on the inertial confinement approach would probably consist of a blast vessel, designed to withstand micro- explosive forces. It would be charged repeatedly with small fuel pellets which would then be irradiated and ignited by short bursts of energy from the driver. To this extent, the reactor cycle might be thought of as resembling the cycle in an internal combustion engine. The reactor, as in the magnetic confinement approach, would also include a blanket for breeding tritium and a coolant for converting fusion-product energy to thermal energy.
