T. KAMMASH,     University of Michigan

ABSTRACT

Recent progress in research towards the development of fusion power is reviewed. In the magnetic approach, the impressive advances made in Tokamak research in the past few years have bolstered the confidence that experimental Tokamak devices currently under construction will demonstrate the break-even condition or scientific feasibility of fusion power. Exciting and innovative ideas in mirror magnetic confinement are expected to culminate in high-Q devices which will make open-ended confinement a serious contender for fusion reactors. In the inertial confinement approach, conflicting pellet temperature requirements have placed severe constraints on useful laser intensities and wavelengths for laser-driven fusion. Relativistic electron beam fusion must solve critical focusing ana pellet coupling problems, and the newly proposed heavy ion beam fusion, though feasible and attractive in principle, requires very high energy particles for which the accelerator technology may not be available for some time to come.

I INTRODUCTION

The fusion nuclear reactions that generate the energy in the interior of the sun and other stars are widely regarded as the most promising means of meeting man’s energy needs for almost all time to come. The vision of a power-producing fusion reactor based on controlled thermonuclear reactions emerged more than a quarter of a century ago, and the remarkable progress that has taken place in recent years has bolstered this vision immeasurably. Although it has been somewhat elusive, the first major milestone towards the achievement of economic fusion power is expected to occur within the next few years when a demonstration of scientific feasibility or break-even is expected to happen.

Unlike most of the other energy sources, the fuel available to fusion reactors is almost limitless. Although it exists in the ocean’s waters to the extent of about one part in 7000, there is enough deuterium in these waters such that if burned by this process it will provide enough energy to meet mankind’s needs for thousands of years and longer. For a net energy to be produced, however, from fusion reactions involving deuterium and tritium nuclei, the fuel must be heated to temperatures in excess of 100 million degrees centigrade (∼ 10keV). At such temperatures the fuel becomes fully ionized leaving behind a collection of negatively charged particles (the electrons) and positively charged particles (the ions) generally referred to as “plasma”. In plasmas of fusion interest the particles move about at very high speeds and collide with one another through the long range Coulomb forces. It can be shown (see next section) that a deuterium (or tritium) ion will travel on the average a distance comparable to the circumference of the earth before it undergoes an energy-producing fusion reaction collision. In fact, a typical ion at these temperatures will undergo a thousand or more scattering collisions before it suffers a fusion reaction collision. To simply place such a hot fuel in an ordinary container is not an effective way of confining the plasma since these particles will collide with the container walls and deposit their energy in them. Not only will this result in melting the walls but, more importantly, it will result in cooling the plasma to temperatures at which the fusion energy production level becomes unacceptable. It has been pointed out by Lawson1 that for a net fusion energy production to take place a deuterium-tritium (DT) plasma at a temperature of 10keV must be sufficiently dense and confined away from material walls for a sufficiently long time to allow enough fusion reactions to take place, with the end result that more energy is produced than consumed. These conditions are generally expressed in terms of the Lawson parameter (nτ) given as the product of the particle density [n (cm−3)] and the confinement time [τ (sec)] which (as we shall see) in the case of DT at 10keV has the value of nτ 1014 cm−3 sec. If we demand that the fusion reactor have a power density comparable to that of other energy sources, such as, fission reactors, then the particle density n must be about 1014, which in turn means that the confinement time (τ) must be about one second. If, on the other hand, we can increase the density to a level which is an order of magnitude (or more) higher than solid state density i.e., 1023, then by the Lawson criterion the confinement time is reduced to a nano-second or 10−9 seconds. In this case energy production occurs on such a fast time scale that it can properly be described as a “microexplosion”. No externally applied means of confinement are needed under these conditions and an “inertial confinement” characterized by the fuel disassembly time will indeed be adequate. This approach is fairly recent when compared to the one envisioned a quarter of a century ago that relies on the use of magnetic fields for the confinement of hot fusion plasmas.

Both of these major approaches to fusion power will be examined in this paper. After a brief section on basic concepts, the main elements of “magnetic” and “inertial” confinement of fusion plasmas will be delineated. The gross characteristics of power reactors based on these approaches will also be analyzed and, finally, conclusions based on present day understanding of the various phenomena and their relevance to the achievement of viable commercial fusion power will be presented.

II BASIC CONCEPTS

In addition to its abundance in nature, deuterium (as well as other hydrogen isotopes) is particularly attractive as a fusion fuel due to the small charge which its nucleus carries. In order for two such nuclei to undergo a fusion nuclear reaction enough energy must first be supplied to overcome the Coulomb repulsive forces. However, since an infinite amount of energy will be needed by classical laws to bring two charged particles into a coalescing contact, quantum mechanical laws allow for barrier penetration at much lower energies due to the wave character of these particles. Putting these two effects together one can write the cross section for such a fusion reaction as

(1)

where E is the energy of the interacting particles in the center of mass system, and A and B are constants that depend on the kinds of particles involved in the reaction. For a DT reaction these constants have the values 2.19×104 and 44.24 respectively(2), and if the energy is expressed in kilo-electron volts the cross section in Eq. (1) is expressed in barns. A plot of this cross section is shown in Fig. 1. Other fusion reactions of interest and their ignition temperatures are given in Table 1. It should be noted that the two branches of the (DD) reaction occur with roughly equal probability while the (D He3) and (pLi6) reactions produce no neutrons. For the purposes of a subsequent discussion we have included in Table 1 the tritium breeding reactions of lithium since tritium does not occur in nature. A quick glance at both Table 1 and Fig. 1 reveals the reason why the (DT) reaction is likely to be the fuel cycle for the first generation fusion reactors inspite of the neutron it produces.

The well-known Lawson criterion alluded to earlier can be readily deduced from a simple energy balance equation which one can write for any volume in the plasma, namely:

(2)

In this equation n1 and n2 refer respectively to the particle densities of the interacting species whose temperatures are denoted by T1 and T2, while τ denotes the energy confinement time. The first term on the left hand side represents the energy produced from fusion carried by the charged particle product, with <σv> denoting the reaction rate. The angular brackets around the reaction rate denote averaging with respect to the particles velocity distribution functions. For Maxwellian distributions the plot of <σv> vs. temperature for various fusion reactions is shown in Fig. 2 where once again we note the distinct advantages of the (DT) reaction. The second term on the left hand side represents the energy loss due to radiative processes (bremsstrahlung and synchrotron) while the right-hand side denotes the thermal energies of the interacting species on the assumption that they possess Maxwell-Boltzmann distributions.

If we employ the familiar results2 for the radiative power losses in the case of DT with equal densities, i.e., (T in KeV),

(3)

and assume that the temperatures of both ion species are the same and equal to the electron temperature, then Eq. (2) will yield an nτ 1014 for the Lawson criterion for the DT reaction at 10keV. Moreover, if we assume that the fusion power produced is just equal to the bremsstrahlung power loss then the resulting temperature is referred to as the “ideal ignition temperature”. Fig. 3 shows such temperature for the...