|
Chapter 3: Various ECF Schemes The details of fuel pellet compression and ignition are very similar in different ICF schemes. The main differences arise in how the energy needed to compress the fuel pellet is generated and delivered to it. Thus, various ICF schemes can be classified according to the nature of the drivers. Fusion-scale lasers and ion-beam accelerators are large and immobile and therefore ill-suited as driver candidates for fusion weapons. However, they contribute to the development of both current generation thermonuclear weapons and pure fusion weapons in several ways, such as by demonstrating the scientific feasibility of ICF, by enabling design of fuel pellets for other schemes, and by allowing more precise computer modeling of other schemes more suited to weapons. Moreover, the various schemes complement each other to some extent. For example, the Heavy Ion Fusion Group at Lawrence Berkeley Laboratory appears to have an ongoing collaborative relationship with Lawrence Livermore's fusion group.1 Table 1 lists the major ICF driver facilities that are either operating or are planned worldwide. The energies and pulse times are provided and where possible supplementary information such as neutron production information is provided. The table does not list all of the facilities which are in early planning stages as well as many smaller operating facilities.
A. Laser Drivers Laser drivers work by first creating a pulse of laser light in the laser medium. 2 A variety of lasing media are possible, each with their own advantages and disadvantages. However, the basic principles are the same. The case of NIF provides a good example of a laser-based ICF facility. NIF's laser system is based upon a neodymium-doped glass laser, the primary type of laser system for ICF research. The 192 beams of NIF actually begin as four pulses of different colors created in the neodymium-doped glass oscillators. These four pulses are split into the 192 beams guided by optical fibers. They are then amplified, have their frequency modulated, and their pulses shaped to match the target requirements. The optical system is complex; the laser pulses pass through a variety of lenses, filters, and amplifiers before being focused onto the target. The result is that the 192 original pulses are amplified from a few nanojoules each to a total energy of 1.8 megajoules. 3 NIF's peak power is expected to be around 500 trillion watts (also called terawatts and abbreviated as TW). Assuming ignition is achieved, the neutron production of NIF will be in the range of 1019 neutrons/shot (assuming a 20 MJ energy output). 4 In comparison, NOVA, a ten beam laser facility at Lawrence Livermore National Laboratory, has a peak power around 120 TW and produces approximately 1013 neutrons/shot. 5 Lasers provide some significant advantages for research into Inertial Confinement Fusion. Laser pulses can be made very short and with high energy, and can be shaped fairly easily. 6 This gives laser fusion facilities considerable flexibility as research tools. However, lasers also have their disadvantages. For facilities such as NIF, design problems include minimizing damage to optical components, creating light of the correct wavelength, minimizing instabilities which affect the symmetry of the compression, and other challenges. The main disadvantages of laser fusion relate to practical applications. For inertial fusion energy one large problem (in addition to those which NIF will have to overcome) is the repetition rate. NIF will only be able to deliver one shot approximately every eight to fourteen hours. 7 By contrast, a large commercial power plant using ICF will require around five shots per second. Laser drivers also have low efficiencies, currently around 1% for solid-state lasers such as those to be used in NIF. Theoretically, this class of lasers can exceed 10% efficiency. However, this would still be lower than the 20% efficiency that ion beam drivers can achieve now. 8 B. Ion Beam Drivers Ions are atoms that are no longer electrically neutral. They have either gained or lost an electron. In the following discussion we will consider positive ions only (atoms that have been stripped of one or more electrons) since they are the ones relevant to ICF schemes. Since the atoms are no longer neutral they are subject to manipulation by electromagnetic forces and can be accelerated to extremely high velocities. There are a variety of electromagnetic acceleration techniques which are described below. In general, the advantages of ion beam approaches are in their high pulse rates and high efficiencies (in comparison with laser systems). Depending on the system, pulse rates are currently approaching or even exceeding the requirements of inertial fusion energy designs. In the case of induction heavy ion accelerators, the pulse rate is high enough that one accelerator may be able to feed multiple beam lines or even reactor chambers. 9 As discussed above, driver efficiency plays a large role in determining the viability of ICF based energy systems. Low efficiencies mean a greater amount of energy is needed to yield a specified driver output. Ion beams have driver efficiencies ranging from 10 to 25 percent, in comparison with around one percent for current laser systems such as those to be used in NIF. 10 Ion beams also have their disadvantages. One problem is that they are difficult to focus. For ion beams to work for fusion, the ion-containing pulses must be extremely short in duration. In other words, the ions must be packed together very closely. However, the positive charges of the ions result in a repulsive force between them. This not only causes the ions to separate and spread out (called "beam divergence"), but can also result in ions reacting with their surroundings during their passage through the accelerator. 11 Achieving high power levels (large energy levels in short periods of time) in an accelerator also poses major difficulties, but specifics depend on the type of accelerator being used. The current requirements for heavy ion beams, consisting of elements such as lead or bismuth, are far lower than those for light ion beams. Heavy ion beam accelerators must be driven at higher energies (1 to 10 giga-electron-volts) in order for the beam to penetrate the target at the proper depth and generate x-rays. This means that the current requirement for a power level of 1,000 terawatts is in the 100 to 1,000 kiloampere range. 12 Light ion beams, consisting of protons or other light ions such as lithium, operate at voltages that are about 100 times lower; the current requirements for a given power level are proportionately greater. 13
 Source: LBL HIF Website 1. Heavy Ion Beams Heavy ions14 can be accelerated using two different types of machines: Induction and Radio-Frequency (RF) accelerators (see below). In either case, an ion source is necessary. A generic ion source would consist of a gas of the desired element being ionized by an electric discharge (such as from a filament). The ions are extracted from the discharge tube by placing a negative electrode outside the ion source to which the positive ions are attracted. 15 a. Induction Accelerators In an induction accelerator the ion beam is accelerated through the use of "pulsers." This figure is only schematic but it provides the basic idea of how the technology works. A doughnut shaped magnet (called a toroid) surrounds the beam. When the circuit is closed and the capacitor discharges, a magnetic field is created in the toroidal magnet. This changing magnetic field creates an electric field in a metal cavity surrounding both the beam and the toroid. These electric field lines accelerate the ions. Many techniques are then used to focus the beam, raise the current, and manipulate the beam in various ways. 16
 Source: LBL HIF Website b. Radio-Frequency Accelerators A radio-frequency (RF) accelerator works on the same basic principle as an induction accelerator; but the method of creating the electric field which accelerates the ions is different. In an RF accelerator the electric field occurs in the gap between successive tube electrodes. The electrodes are fed by a radio-frequency source. The alternating current of the source means that the electrodes continuously switch back and forth from positive to negative. Since successive electrodes have opposite charges, an electric field is created in the gaps. The accelerator is designed so that ions pass through the electrode gaps at exactly the correct times so as to be accelerated by the electric field in pulses which are coordinated to occur when the field lines are in the direction of beam travel. 17 While RF accelerators are a well-established technology, they face certain limitations. Aside from expense, the RF accelerator by definition is tied to the frequency of its RF generator, which can be doubled or tripled, but is essentially limited. Since the beam current is proportional to the frequency, this imposes limitations on the beam current. 18 Heavy ion beam accelerators have some significant advantages as drivers for ICF energy schemes. Indeed, it seems to be the prevalent view in the literature that while laser drivers are necessary for the experimental and demonstration phases of target development, ignition, and other developmental research, heavy ion beams would be used if fusion energy were ever commercialized. The main advantages of heavy ion beams are their long life-times and fast repetition rates. Heavy ion beams can achieve their peak power levels with relatively lower current levels, reducing some of the development problems as compared to light ion beams which require currents significantly higher than now achievable. A commercial fusion plant would probably need to explode around 5 to 10 D-T capsules a second for many years. Heavy ion linear accelerators (linacs) have pulse rates of 10 to 1000 pulses per second. 19 Therefore, the power plant would not be limited by the repeatability of the driver. By contrast, lasers and some light ion beams require far longer times between pulses. For instance, the maximum rate for NIF is projected to be 600-1200 per year. On the other hand, it remains to be demonstrated whether heavy ion accelerators can reach the high peak power necessary for fusion and be able to adequately focus the beams on small targets.
 Source: LBL HIF Website 2. Light Ion Beams Light ion beam accelerators use a single acceleration gap to achieve the necessary particle energies (~10-30 MeV). This acceleration gap consists of a negative and positive electrode (cathode and anode) which are supplied with pulse power. Unlike heavy ion accelerators, the ion source is part of the accelerator, since the lithium (or other light ion) source forms part of the anode material. The light ion beam hohlraum would be constructed of a high Z material outer wall and filled with a carbon-based foam. Unlike laser hohlraums, however, the ions are not converted into X-rays by the wall material, but rather by the foam fill. The ions pass through the thin wall of the hohlraum and the foam converts their energy to soft X-rays. The fusion target is then uniformly irradiated by the X-rays, which are now contained by the high-Z hohlraum wall. This means that the target must be carefully designed so that the energy is deposited at the proper depth in the target. 20 Like heavy ions, light ion beams have the advantage of higher efficiencies than laser systems. Current light ion beam efficiency is around twenty percent. 21 However, there are still significant gaps between current performance and the requirements for commercial power from inertial fusion (called IFE or inertial fusion energy).
A number of practical difficulties prevent light ion beams from being closer to the requirements of fusion power. The diode ion source must produce a beam of uniformly charged ions of a single type. This must be done at high power levels and efficiencies. Some problems are similar to those for heavy ion beams. The ion beam must be focused on a small spot. This limits the size of the initial beam radius and therefore, the size of the acceleration gap used to extract the ions. A long acceleration gap results in higher energy ions, but the ions have a longer distance over which to spread out, increasing the spot size. In order to focus the beam, various problems of beam divergence must be overcome (some of which result from the positive charges on the ions). A further problem with light ion diodes is the creation of significant numbers of free electrons due to the large electric field. These free electrons can prevent high currents by neutralizing ions (a potentially significant problem considering that the light ion beam approach requires high beam currents). Magnetic fields are often applied to minimize the flow of electrons. 24 Repetition rates also need to be improved for light ion systems. As of 1995 Sandia's HERMES III pulse power system could operate at about seven pulses per day. A repetitive system was in the test phase which had managed to reach 120 Hz for 18 minutes, although at a slightly lower energy and current. 25 However, this indicates that research towards a useable repetitive pulsed power system for light ion beams is progressing. 26 Thus, there are still significant advances that need to be made in light ion beam and target technology in order to achieve the proper energies and powers while overcoming problems such as beam divergence. C. Z-pinch While useful for fusion research ion beams and lasers cannot function as drivers for pure fusion weapons. However, a pulsed power device known as the "wire-array z-pinch" has this potential. The name of the device derives from the fact that it is a cylindrical array of wires (the vertical direction of a cylinder is usually denoted by the letter "z", for z axis) and the fact that the cylinder is "pinched" to a very small diameter. The potential of the device arises from its already established capacity to generate x-rays at energy levels significant for pure fusion explosions and from the possibility that it could be miniaturized. Significant improvements in the wire-array z-pinch have occurred at the Sandia National Laboratory over the past few years, where a device called the Particle Beam Fusion Accelerator-Z (PBFA-Z) has reached levels that had previously been thought to be unattainable. In particular, recent laboratory reports state that PBFA-Z has generated 2 MJ x-rays, a level comparable with that planned for the National Ignition Facility. In the z-pinch wire-array experiments a large current is passed through a large number of very thin wires arranged in a cylindrical bundle. As the current rises, the magnetic field associated with it increases. This in turn compresses the array of wires into a cylinder of progressively smaller diameter. At the same time, the high current is rapidly heating the wires, evaporating the wire material, and turning it into a plasma. As this plasma is compressed further by the magnetic field, the electrons and ions forming the plasma come to an abrupt stop (this is called stagnation). This abrupt stop converts the kinetic energy of the particles into x-rays. The process is somewhat analogous to the conversion of the kinetic energy of a car into heat during sudden braking. Since x-rays can be used to compress a fusion fuel pellet, the high level of x-ray energy achieved by the wire-array z-pinch makes it very interesting to fusion researchers. The initial energy source for the z-pinch experiments at Sandia was a pulsed-power generator used for light ion research. This apparatus was called PBFA (Particle Beam Fusion Accelerator). A large capacitor bank was used as the electrical energy source. By September of 1996 this capacitor bank was converted for use as the energy source for the wire array z-pinch experiments discussed above. 27 The recent performance levels announced for PBFA-Z (290 TW) demonstrate the potential of this technology. The experiments have exceeded most of the milestones that have been set in a relatively short period of time. Sandia National Laboratory has officially requested permission from DOE to design the next generation of x-ray facility, the X-1. While no official design has been produced, there are articles indicating that conceptual designs have been completed. These indicate that X-1 would produce x-rays of approximately 16 MJ. 28 D. Chemical Explosives Chemical high explosives (HE) are an integral component of current nuclear weapons, since they trigger the fission primaries of these weapons. However, the requirements of high explosives for current nuclear weapon use are much less stringent than those for pure fusion weapons. In the latter case, they would have to meet performance requirements similar to those of other fusion drivers. They must not only deposit sufficient energy into the device, but also be powerful enough to compress the fuel to fusion densities and temperatures fast enough to avoid premature disassembly, and be uniform enough to avoid instabilities. High explosives in pure fusion weapons would likely have to be used in combination with other techniques. The two major problems with using chemical explosives alone as fusion drivers are their comparatively low energy densities and their slow detonation velocities. We explore them here because their use in combination with electromagnetic approaches to plasma compression could be crucial to miniaturizing pure fusion devices. 29 High explosives are complex molecules generally consisting of carbon, hydrogen, nitrogen, and oxygen. The generic abbreviation for these explosives is CHNO. CHNO explosives can also contain other elements, such as fluorine. The energy release from explosives occurs by a process called oxidation. A variety of chemicals are produced during this oxidation process, including nitrogen, water vapor, carbon monoxide, and carbon dioxide. The amount of each element in the starting explosive will determine how much of each product is formed (largely depending on the availability of oxygen) and whether the explosive is under- or over-oxidized. When the explosive is exactly oxygen-balanced, it will have the highest energy density (energy per unit weight) possible for that type of explosive. TNT, for example, is under-oxidized. In other words, it is not as efficient an explosive as it could be. The composition of an explosive also has a bearing on its detonation velocity (see below). 30 The relatively low energy density of current high explosives limits their use as drivers for pure fusion weapons. High explosives have energy densities in the range of 5-6 kilojoules (kJ)/gram. The energy necessary to ignite the core of a 1 milligram D-T fuel pellet is on the order of 10 kJ. 31 Assuming a one percent efficiency in the coupling of the energy in the explosive to the fuel pellet, about one megajoule of explosive energy would be required, amounting to about 200 grams of high explosive. This poses severe physical problems because the volume of the explosive (on the order of 100 to 150 cc) would be more than five orders of magnitude larger than the volume of the D-T fuel that it is supposed to ignite(about 0.005 cc). 32 This makes the need for a fast and efficient coupling between the release of explosive energy and the fuel pellet a central problem in the use of high explosives in pure fusion weapon development. At the same time, high explosives are far more compact than lasers, ion beams and other energy storage devices that are used in ICF research, a key factor making them more favorable for fusion weapons use. Much of the practical problem of creating pure fusion weapons can therefore be viewed as the exploration of ways in which the energy of high explosives can be transformed so as to create a sufficiently efficient and rapid coupling with a relatively small fuel pellet. For instance, this is the central idea in the use of high explosives in magnetized target fusion experiments. E. Advanced materials manufacturing Advanced materials manufacturing may radically improve the prospects for pure fusion weapons by possible making smaller, more efficient, more precise, and less costly components such as drivers and ablators (the outer layer of the fuel pellet which is evaporated). A variety of advanced manufacturing techniques may affect the field. Development of new materials, lasers, explosives, electrical devices, and other components of explosive confinement fusion devices may also dramatically alter the prospects of developing these weapons. The consequences of radically new technologies, materials, and manufacturing processes are notoriously difficult to predict, and we do not attempt to do so here. The purpose of this section is simply to point out that emerging processes and materials may substantially and rapidly increase the feasibility of pure fusion weapons. Just as it would have been impossible to forecast the present state of the Internet and personal computers from the vacuum tube era of the 1940s, we cannot accurately predict where fusion technology will be in the next decades given the continued tremendous pace and variety of technological change. We will illustrate the possibilities by discussing a few processes and technologies that may have particular bearing on the development of pure fusion weapons. For instance, a reduction in the size of capacitors by an order of magnitude appears possible with technologies now being developed by the Pentagon. New manufacturing techniques and improved dielectric materials could combine to make possible capacitors with energy densities on the order of 10 joules per cc. 33 Thus, one megajoule of driver energy could be stored in a volume of 0.1 cubic meter. This is still far greater than the volume for the same energy in a chemical explosive, yet it is small enough to enable a wire-array z-pinch device using a few milligrams of D-T fuel to be portable. Development of efficient coupling of both capacitor stored energy and chemical drivers, via techniques such as magnetized target fusion, could result in practical pure fusion weapons. Based on current projections of fusion yield from MTF, Jones and von Hippel calculate a total yield of 0.5-2.5 tons of HE for a device weighing three tons. 34 1. Nanotechnology The development of manufacturing by precise manipulation of small numbers of molecules or even single atoms provides another example of new techniques whose potential is not possible to project at present, but which may have substantial impact. The approach goes under the rubric "nanotechnology," which means technology operating at a scale of one-billionth of a meter. This is thousand times smaller than the micro-scale technologies that gave us the computer chip. Nanotechnology may have substantial implications for pure fusion weapons, ranging from development of improved explosives to more efficient ablators. Nanotechnology is a relatively new, but rapidly growing, field combining physics, chemistry, material science, and engineering. K. Eric Drexler defines molecular manufacturing, a goal of nanotechnology, as "the construction of objects to complex, atomic specifications using sequences of chemical reactions directed by non-biological molecular machinery." 35 This would involve synthesis of a fundamentally different nature than current methods of chemical synthesis. Atoms and molecules would be guided to react with one another in a highly controlled fashion at the individual molecule level. Therefore, unlike conventional synthesis, the reactions would not be dependent on collisions proportional to reagent concentrations, spatial effects, and electronic interactions between reagent molecules. Instead, reactions would result from proper positioning of individual reagent molecules. 36 Precise positioning (albeit using macroscale instruments rather than nanoscale molecular machines) of atoms and molecules has been demonstrated. One potentially very important application of nanotechnology to the development of pure fusion weapons is in the development of chemical explosives. The current process of developing high explosives begins with a theoretical exploration of possible candidates for a new highly energetic material. Both theoretical and synthesis chemists (who would be responsible for making the explosive) determine candidate molecules worth further exploration. Using powerful computers, the candidate molecule's shape and binding energy are modeled and its explosive properties (e.g. detonation velocity, energy) are predicted. Lawrence Livermore scientists are already taking advantage of the capabilities of the Accelerated Strategic Computing Initiative (ASCI) program for advanced modeling capabilities and these capabilities can be expected to expand as the ASCI program is further developed. The next step after computer modeling (assuming the models show the candidate molecule is worth pursuing) is to synthesize a small amount of the explosive. This appears to be one of the most difficult steps, since the material has never been synthesized before and the chemists must begin from scratch using a trial and error process (although based upon previous experience). In some cases, the necessary reagents to produce the explosive are too expensive, harmful to the environment, or dangerous. If it is possible to synthesize the new explosives, small quantities are made for laboratory testing. These tests not only determine its explosive parameters, but also test for its safety, stability against degradation and other factors. According to a Livermore article on chemical explosive research at the laboratory, the potential safety problems are the most common cause of rejecting candidate materials. The final stages of explosive development are to mix the explosive with other materials (the "formulation" stage) and to scale up the process to production levels. The formulation step also requires a certain amount of trial and error. At the end of the development process only a fraction of the possible explosives are actually developed. The rest are discarded for reasons such as difficult synthesis, poor performance, or safety. As the Livermore article states, "Developing new energetic materials is a complicated process in which many candidate molecules are considered, a few synthesized, even fewer formulated, and only a small handful adopted by the military or industry." 37 The problem of synthesis seems to be a particularly vexing one for the explosives industry. Livermore can create a computer model of a candidate explosive in about a week. However, it may take a year or more for the chemist to develop the right synthesis scheme. 38 Chemical synthesis of explosives, like most chemistry, is done in what is called the gas or liquid (also called "solution") phase. This means that the chemicals being mixed, heated, stirred, etc. are either gases or liquids. The molecules in each reagent react with the molecules in the other reagents, essentially in a statistical fashion (i.e. by random collisions). The two reagents mix, the molecules move around one another, bumping into each other, and in some cases reacting with one another to form a new molecule. As one scientist noted, "traditional manufacturing methods spray atoms about in great statistical herds." 39 This process can be aided by various techniques, such as applying heat to increase the reaction rate. This must be regulated since too great a temperature can result in degradation of the reagents. This new molecule must then be extracted if it is to be used. The synthesis scheme must include a number of intermediate reactions before the final product is synthesized. The synthesis process is an inherently inefficient one. The resulting product is never produced at 100% efficiency, the existence of impurities can be a problem for both the reaction and the final product, and other reactions can take place producing unintended or undesired side-products. Furthermore, as discussed above, it may not always be possible to synthesize the desired product due to reagent unavailability, toxicity, cost, and other reasons. The ability to manipulate single molecules using nanotechnology may affect the development of chemical explosives in two ways. On a manufacturing level, explosives could be manufactured more easily and closer to their theoretical maximum density. This comes from moving away from solution-phase chemistry in which reagents are mixed to get a certain yield of end-product. More significantly, however, nanotechnology could open up the possibility of a new generation of explosives. As discussed above, explosives development is not always constrained by what is theoretically possible, but by what is practically possible in the laboratories trial and error process of synthesizing candidate materials. It may not always be clear how a new energetic material could be synthesized, the reagents may be too costly or dangerous, or the number of synthesis steps may not be practical for full-scale manufacturing. 40 Molecular manufacturing might be able to eliminate those problems by relying on raw materials that are easier to acquire and use. In general, reagents must be able to react with other molecules in order to form a new product. The problem in conventional synthesis is that a reagent which too readily reacts with other molecules will result in unwanted reactions. Therefore, reagents must be chosen which will be selective in their reactions with other molecules. In the case of molecular manufacturing, as stated above, highly reactive reagents will be able to be controlled and their reactivity can become an asset (potentially increasing reaction frequencies). 41 2. Metallic Hydrogen Another development which may have significance in the future for the development of pure fusion weapons is the reported, but as yet unconfirmed, experimental discovery of metallic hydrogen. The higher density of metallic hydrogen may provide benefits for the design of fusion weapon capsules (and may reduce the amount of fuel necessary for larger pure fusion weapons). 42 Metallic hydrogen was first theoretically postulated in the 1930s. Since then there have been a variety of attempts to achieve the high pressures and other conditions necessary for hydrogen to become a metal. Various researchers have claimed success, but have had their discoveries overturned upon further experimentation. 43 Most recently, a group of researchers at Livermore claimed to have made metallic liquid hydrogen in a shock experiment using a gas gun. 44 While this experiment has not been repeated by other laboratories, there appears to be some evidence that the hydrogen was metallicized, although for a very short period of time. 45 The Livermore experiments could also have a significant impact on the study of fusion and pure fusion weapons. The improved understanding that these experiments provide of the behavior and properties of hydrogen (and its isotopes) at high pressures and temperatures could aid fusion scientists in tuning their lasers, improving their computer codes, and designing better targets for NIF. This could result in higher fusion energy yields and make the NIF target performance range "broader and more flexible." 46 This would occur whether or not metallic hydrogen were ever brought to a stable room pressure and temperature form. However, if it were possible to produce metallic hydrogen in a useful form, then the implications for fusion weapons could be greater. Metallic hydrogen may also be an extremely powerful explosive, releasing large amounts of energy. 47 However, this is at present speculative, given the uncertain state of metallic hydrogen research. It is unknown at this time whether metallic hydrogen could ever be produced in a useful form.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Institute for Energy and Environmental Research