Approaches to Fusion Energy

Historically, two approaches to harnessing fusion power for energy production have been followed. The first is magnetic fusion energy (MFE) which creates "magnetic bottles" that need to hold the plasma for a time of order one second (Chen, 1974; Sheffield, 1994). The best known example is the tokamak, a toroidal geometry. (A toroid is a donut shaped configuration, which is preferred for most magnetic fusion applications, because magnetic field lines can encircle the toroid nearly endlessly without intercepting a wall. Plasma ions and electrons flow along magnetic field lines much more easily than they flow across them, so plasma loss rates to walls are reduced in toroids.) The second approach is inertial fusion energy (IFE) which today uses powerful lasers, which compress a millimeter-sized capsule of fusion fuel to the high densities and temperatures needed for fusion to occur (Hogan, 1995). For IFE, the confinement time is the time for the capsule to blow itself apart, of order nanoseconds (10~9 s): hence the name "inertial" fusion. These approaches are at roughly similar stages of scientific development: magnetic fusion has demonstrated up to 60% as much fusion power produced as heating power injected into the plasma in the JET (Joint European Torus) tokamak (Watkins, 1999). Even better confinement has been demonstrated in the JT-60 tokamak in Japan, but without the use of deuterium-tritium fuel, resulting in much less generation of fusion energy (Ishida, 1999). The inertial fusion program is constructing the NIF, National Ignition Facility, with a goal of demonstrating ignition and energy gain of ten near 2100 (Kilkenny et al., 1999). Ignition means that the 3.5 MeV alpha particles from fusion reactions deposit enough energy in nearby fuel to heat it to fusion energies, analogous to heating a log so that it ignites and burns. Energy gain is the ratio of fusion energy out to energy in.

A generic fusion power plant, Figure 8.1(a), consists of a central reaction chamber (the core), surrounded by an approximately 1-m thick blanket that contains lithium. The blanket performs two functions: it generates the tritium fuel from neutron interactions with the lithium, and it shields the surrounding structure from neutron bombardment. Penetrations through the blanket must

Figure 8.1 (a) Generic fusion core with blanket/shield. (b) Toroidal fusion device. Courtesy of the University of California, Lawrence Livermore National Laboratory, and the Department of Energy under whose auspices the work was performed.

be provided for injecting fuel and energy for either MFE or IFE. These may be from one side as shown, but more generally are from opposite sides or even uniformly distributed around the periphery. Outside the blanket are superconducting magnets for magnetic fusion, and beamlines for inertial fusion. The power plant in Figure 8.1(a) could be spherical or cylindrical. The other major geometry studied is the torus shown in Figure 8.1(b).

8.3.1 Magnetic fusion energy

Many concepts in magnetic fusion have been studied during the last four decades, with a strong emphasis on the tokamak during the last two decades. The tokamak consists of an externally applied toroidal magnetic field Bt, and a toroidal plasma current to create a poloidal magnetic field, Bp, Figure 8.1(b). A transformer drives the plasma current in nearly all tokamaks, with auxiliary current drive by radio-frequency waves or neutral-atom beams on many toka-maks. Additional externally applied magnetic fields provide vertical fields for plasma-position control and shaping, and can provide divertor geometries, where the outermost field lines strike a divertor plate to allow the removal of impurities from the edge plasma. Within the last decade, innovations in toka-maks have dramatically increased their ability to confine plasma. The ARIES-RS, Figure 8.2, applies these innovations to improving the power plant core (Najmabadi, 1998).

Tokamaks keep refusing to be written off as too large and complicated, and have reached performance levels that would allow the construction of a burning plasma experiment. (A burning plasma is one in which the fusion power in alpha particles heats the plasma at a rate exceeding the external heating power.) Potential experiments under consideration encompass short-

Figure 8.2 ARIES-RS Tokamak power plant. Courtesy of Farrokh Najmabadi, University of California, San Diego.

pulse (few seconds) machines with normal-conducting magnets and long-pulse (thousands of seconds) machines with superconducting magnets. An example of the former is FIRE (Fusion Ignition Research Experiment) (PPPL, 2000) with a price tag of around $1-1.5B. The latter type of machine is exemplified by ITER (the International Thermonuclear Experimental Reactor) (Shimomura, 1999) under study by the international fusion community. ITER would be a full engineering test reactor capable of testing engineering components and, accordingly, carries a price-tag of —$5-10B depending on performance requirements. Such burning plasma experiments will enable many of the remaining physics issues for a power plant to be studied. However, the present lack of low-activation materials prevents the conventional tokamak from being on an environmentally attractive power plant path. At present, therefore, it should be considered as a physics and technology experiment, not a power plant prototype.

The stellarator is a toroidal magnetic configuration in which all confining magnetic fields are imposed by external coils. It has the advantages over a tokamak of not needing current drive, and not disrupting (in a disruption, the

Figure 8.3 Spheromak configuration has simpler geometry, that is more compatible with liquid walls, than devices with toroidal blankets. Courtesy of the University of California, Lawrence Livermore National Laboratory, and the Department of Energy under whose auspices the work was performed.

plasma current decays within a few tens of milliseconds, much faster than usual, and exerts large forces on the walls). Stellarator coils are more complicated; but, in the modular type, the multiple versions of potato-chip-shaped coils are not trapped or linked by other coils and so can be more easily disassembled for maintenance. As a power plant, it shares most of the other characteristic of a tokamak, and is expected to yield a similar cost-of-electricity.

Other magnetic configurations are described as self-organizing, because more of the currents generating the magnetic fields are carried by the plasma itself (Riordan, 1999). For example, the spheromak has both toroidal and poloidal currents to create both the poloidal and toroidal magnetic fields respectively. A spheromak plasma will persist for some time when placed inside a can with conducting walls to carry the image currents. The pot blanket spheromak, Figure 8.3, is very simple, but uses a metal wall to separate the liquid from the plasma; this wall will limit the power density and will be damaged and activated by neutrons. In other concepts, a swirling liquid is against the plasma. Another magnetic configuration is the field-reversed-configuration (FRC)

which generates only a poloidal field with plasma currents, and which is immersed in an axial magnetic field.

We have just classified magnetic bottles by the degree to which the magnetic field is created by currents in external coils versus currents flowing in the plasma. Another useful classification criterion is the topology of the chamber: a cylinder, sphere, or toroid; and if the latter, the amount of structure required in the central hole of the toroidal geometry. The minimum size of the fusion core of a power plant is greatly increased by such structure, because it generally includes superconducting magnet windings that must be shielded against 14 MeV neutrons. Neutron shields are about 1 m thick, Figure 8.1, to reduce the refrigeration requirements at 4 K, as well as to reduce neutron damage, thereby extending the life of the windings. Therefore, the central hole must have a minimum diameter of about 2 m, and small, high-power-density fusion cores are not possible with such geometries.

Some toroidal concepts, such as the spherical tokamak, avoid this restriction by using unshielded room-temperature magnet coils in the center leg. This allows the center leg to be much smaller than one meter in diameter. In determining the attractiveness of the spherical tokamak, one needs to evaluate the impact of the circulating power to drive the center leg current, and the economic and environmental consequences of regularly replacing the activated center conductor.

We now use these classifications to further discuss two potentially attractive examples of confinement configurations. A more compact structure with no structure in the center, such as the spheromak or the FRC, would provide a lower-cost development path, lower cost-of-electricity and smaller volume of activated material, if adequate confinement could be achieved. Our present knowledge of confinement properties is poorer for this class. The fusion program today is building small experiments that will test these concepts with better power plant potential - to determine whether they can be made to confine plasma sufficiently well. This broadens the policy of the last 15 years of developing only the tokamak because it has the best confinement and is best understood, so that its confinement and performance at power plant scale can be confidently predicted.

The reversed-field pinch (RFP) is also largely self-organized, but requires magnet windings within the donut hole. It therefore requires a larger fusion core and looks more like a tokamak, except for a lower applied toroidal magnetic field. Although RFPs may have less potential as power plants, they have proven to be very productive in increasing our physics understanding of these self-organized configurations (Prager, 1999).

The spheromak and FRC are the MFE configurations most likely to be compatible with thick liquid walls; because their simpler boundary shape makes formation of liquid walls less difficult, and their high plasma density is more likely to withstand the high vapor pressure of the liquids. As with many configurations, spheromaks gain stability from the plasma pressure against the walls; whether liquid walls are too easily pushed aside by the plasma remains to be determined. Even if steady-state solutions do not exist, pulsed configurations may be attractive, if their repetition rate is high enough to minimize thermal cycling in the liquid that operates the steam cycle (Fowler et al., 1999).

The extreme pulsed limit of magnetic fusion is magnetized target fusion, MTF (Drake et al., 1996), which fits a largely unexplored niche between MFE and IFE. MTF employs magnetic fields to reduce electron thermal conduction, thereby allowing plasma densities orders of magnitude lower than with IFE, but with plasma density and magnetic fields exerting pressures too large to be restrained in a steady-state device. As a result, some of the structure must be replaced each shot. The structure to hold the plasma is small and relatively inexpensive, allowing multiple confinement geometries to be evaluated in one facility. The development path to ignition may be the least expensive of any fusion concept. The path to a power plant is less clear, but might involve replaceable structures of frozen liquid and/or closely coupled liquid walls, fusion yields exceeding a gigajoule, and repetition rates of slower than one per second, to allow time for reloading.

8.3.2 Inertial fusion energy

Inertial fusion (Hogan, 1995) requires depositing energy in a short time (less than 10 nanoseconds - the time for light to travel 3 m or 10 ft.) on the outside of a millimeter-sized capsule. The outside of the capsule blows off at high velocity, causing the rest of the capsule to rocket inwards. This compresses the fuel to high density and temperature, with the goal of reaching sufficiently high values to ignite a fusion burn.

Inertial fusion is characterized as direct- or indirect-drive, Figure 8.4. For direct drive, the energy to compress the capsule of fusion fuel is incident directly on the outside of the capsule. Experiments with lasers have demonstrated that the energy must be extremely uniform over the surface of the capsule in order to have the entire surface rocket inwards at the same velocity, a requirement if the capsule is to remain spherical during compression. Variations in velocity from nonuniform illumination drive large amplitude waves, the so-called Rayleigh-Taylor instabilities on the surface. Such waves prevent the capsule from compressing as a sphere by the 10-30 fold in radius that is required to obtain fusion energy.

Fusion Approaches Laser Sphere

Figure 8.4 Direct drive energy is directly incident on the outside of the capsule. Indirect drive energy is deposited on the walls of a hohlraum, which heats up and emits x-rays that illuminate the capsule. Either can be driven by lasers or ion beams. Courtesy of the University of California, Lawrence Livermore National Laboratory, and the Department of Energy under whose auspices the work was performed.

Figure 8.4 Direct drive energy is directly incident on the outside of the capsule. Indirect drive energy is deposited on the walls of a hohlraum, which heats up and emits x-rays that illuminate the capsule. Either can be driven by lasers or ion beams. Courtesy of the University of California, Lawrence Livermore National Laboratory, and the Department of Energy under whose auspices the work was performed.

For indirect drive, Figure 8.4, the energy is deposited on the inside of a can or hohlraum (German for "radiation-room"), which contains the capsule. The energy creates soft x-rays of a few hundred electron volts (few million Kelvin) energy, that reflect off the walls of the can and uniformly illuminate the capsule. This is analogous to indirect lighting, that creates a relatively uniform, diffuse illumination, as contrasted with direct illumination which is more efficient but creates a higher contrast lighting. Direct drive is similarly more efficient, since energy is not expended in heating the inside of the hohlraum. Indirect drive is more compatible with liquid-wall fusion chamber design, because the beams can be clustered to come in from one or two sides, rather than distributed uniformly over the surface of a spherical chamber.

Inertial fusion is also categorized by the type of driver it uses. Experiments to date have used lasers; most of these are flashlamp-pumped solid state lasers that have low electrical efficiencies (of order 10~3 to 10~2) and low repetition rates of ~ 10 shots per day. A few experiments have also used z-pinch radiation sources or light ions. Other drivers are being developed. DPSSLs (diode-pumped solid state lasers) should deliver efficiencies near 10%, and repetition rates of 10 Hz. Electron-beam pumped gas lasers, such as KrF (krypton

Figure 8.5 HYLIFE-II heavy-ion driven, inertial fusion power plant. Plant layout shows the driver, target and chamber. The use of liquid jets to protect the fusion chamber results in long lifetime, low cost, and low environmental impact. Courtesy of the University of California, Lawrence Livermore National Laboratory, and the Department of Energy under whose auspices the work was performed.

Figure 8.5 HYLIFE-II heavy-ion driven, inertial fusion power plant. Plant layout shows the driver, target and chamber. The use of liquid jets to protect the fusion chamber results in long lifetime, low cost, and low environmental impact. Courtesy of the University of California, Lawrence Livermore National Laboratory, and the Department of Energy under whose auspices the work was performed.

fluoride), have nearly as high efficiency. Heavy-ion accelerators are attractive driver candidates for inertial-fusion energy, building on extensive experience with high-energy and nuclear physics facilities. They promise high efficiency, high repetition rate, long life, and magnetic final optics to focus the beam onto the target (Bangerter, 1999). Compared with the mirrors or lenses that lasers need, magnetic optics are relatively immune to the effects of target explosions, because neutrons, x-rays, and debris can pass through the aperture of the magnet, while the magnet windings are shielded. The high beam currents required with heavy-ion fusion are a new and challenging element for both accelerators and focusing systems. Target designs have been developed for both laser and ion drivers, as indicated in Figure 8.4, which are predicted with 2-D codes to have reasonable gains (near 100) for a power plant.

A heavy-ion driven, inertial fusion power plant is shown in Figure 8.5. It is characterized by modularity, where the driver, chamber, target fabrication plant, and target injector are all separated. Thick liquid walls within the chamber, composed of jets of Flibe, enclose the reaction region, and protect the solid chamber walls from neutrons and shock waves.

IFE provides a paradigm shift from MFE tokamaks in two areas. First, it places most of its complexity in the driver, which is decoupled from the fusion chamber, whereas MFE incorporates most of its complexity immediately around the fusion chamber. This provides IFE with potential advantages of (a) reducing development costs by allowing the chamber and driver modules to be upgraded independently, and (b) reducing operational costs by delivering a higher availability to utilities. Higher availability is possible because the complexity within the fusion chamber is minimized, and repairs to systems outside the neutron shield can be made more rapidly, in some cases even while the driver continues to operate (Moir et al., 1996). Second, indirect-drive IFE (heavy-ions, or possibly lasers) offers the potential for lifetime fusion chambers with renewable liquid coolants facing the targets, instead of solid, vacuum-tight walls that could be damaged by heat and radiation. Protected in this way, all of the chamber structural materials would be lifetime components. Their minimal residual radioactivity would mean that at the end of the fusion plant's life, most of the materials could be recycled, rather than requiring deep underground disposal.

However, much of the technology and engineering capability for conventional tokamaks is near the levels required for a power plant. This is not the case for IFE, nor MFE alternative concepts, where today's experiments are much further from power-plant technology. A consequence of the philosophy presented here, is that in seeking more attractive concepts for fusion, we push fusion energy further into the future. We believe that further into the future is preferable to never. "Never" is the likely result of trying to build power plants on which the public is not sold.

A concept that could dramatically reduce the driver energy is the fast ignitor (Tabak et al., 1994). With conventional IFE, the capsule is heated by a high degree of compression to ignite at the center, Figure 8.6a. With the fast ignitor, the capsule is compressed to moderately high density but low temperature. A portion of it is then heated to ignition by a very short energy pulse with duration measured in picoseconds (millionth of a millionth of a second). The hot spot heats the surrounding region, and the fusion burn propagates through the fuel, Figure 8.6b. This concept can give gains (the ratio of fusion energy out to driver energy in) of a factor of 5 to 10 higher than conventional IFE. Relaxing the compression ratio allows a lower energy and therefore less expensive driver, resulting in a lower cost-of-electricity (COE). However, several difficulties must be overcome first: the driver must bore a hole through the plasma surrounding the capsule, then deliver the fast-ignitor pulse through the hole and focused to

Figure 8.6 Inertial confinement fusion targets using the "fast ignition" concept, showing laser or heavy-ion driver beams (arrows) and fast ignitor laser focused on the target from the left. The illumination geometries are (a) direct drive symmetrically illuminated and (b) indirect drive, illuminated from one end. In (a) the ignition laser has to penetrate an overlying plasma corona before heating the target. In (b) using cone focusing within a hohlraum, the plasma blowoff is excluded from the ignitor laser path and the driver lasers can be concentrated at one end of the chamber. Courtesy of the University of California, Lawrence Livermore National Laboratory, and the Department of Energy under whose auspices the work was performed.

Figure 8.6 Inertial confinement fusion targets using the "fast ignition" concept, showing laser or heavy-ion driver beams (arrows) and fast ignitor laser focused on the target from the left. The illumination geometries are (a) direct drive symmetrically illuminated and (b) indirect drive, illuminated from one end. In (a) the ignition laser has to penetrate an overlying plasma corona before heating the target. In (b) using cone focusing within a hohlraum, the plasma blowoff is excluded from the ignitor laser path and the driver lasers can be concentrated at one end of the chamber. Courtesy of the University of California, Lawrence Livermore National Laboratory, and the Department of Energy under whose auspices the work was performed.

a very small size, at precisely the right time, with tolerances of less than 100 picoseconds (100 X 10~12 s).

The fast ignitor in IFE and the several new MFE concepts mentioned are examples of innovations that may significantly increase the attractiveness and lower the cost of electricity from a fusion power plant. Even nearly 50 years into the fusion program, new ideas continue to emerge; and old, previously rejected, ideas become attractive with new technology.

Scientific understanding of inertial-confinement fusion will be significantly advanced early in the next century by the completion and operation of the NIF (National Ignition Facility) at Lawrence Livermore National Laboratory in the US, and the Laser MegaJoule (LMJ) facility in France (Kilkenny). The NIF is likely to be the first laboratory device to realize fusion ignition - when its remaining issues, that currently cloud the cost and schedule for achieving ignition, are solved (Malakoff and Lawler, 1999). Although the primary missions of both the NIF and the LMJ are defense related, an important spin-off benefit is to show that inertial-fusion energy is feasible. To capitalize on this demonstration requires a parallel effort in developing suitable and cost-effective drivers and chambers for IFE.

Was this article helpful?

0 0
Solar Power

Solar Power

Start Saving On Your Electricity Bills Using The Power of the Sun And Other Natural Resources!

Get My Free Ebook


Post a comment