Institute for Energy and Environmental Research
|
Chapter 4: The Prospects for Pure Fusion Weapons There are currently two approaches to fusion being researched that could lead to pure fusion weapons, or more broadly, to nuclear weapons that do not require a fission trigger. They are inertial confinement fusion (ICF) and fusion driven by various combinations of electrical, electromagnetic and chemical compression of plasmas, such as Magnetized Target Fusion (MTF). Neither of these technologies is sufficiently developed to demonstrate the scientific feasibility of these weapons. But the research paths and the stated goals for both of them are such that, if successful, the prognosis for such weapons could change dramatically. None of the projects have the development of pure fusion weapons as their officially stated goal. In view of the many commitments that the nuclear weapons states have made to stop developing new nuclear weapons, most recently as part of the Comprehensive Test Ban Treaty, these same states could hardly announce that they are developing radically new nuclear weapons. The questions then revolve not around stated intentions, but around the technical capabilities that the pursuit of high power ICF and MTF programs will give the nuclear weapons states, and in particular the United States, France, and Russia. If the technical potential for building these weapons is developed, or even if their scientific feasibility is established, the pressures to build them, especially in times of crisis, would be immense. For these reasons our evaluation of these technologies and their implications for nuclear disarmament and non-proliferation focuses on the development of technical capabilities. In this chapter, we will examine the technical goals of these projects as they relate to the requirements for building pure fusion weapons and to a lesser extent non-fission triggered nuclear weapons. A. Requirements for pure fusion weapons Fusion weapons that do not need a fission trigger have been considered mainly for two military and technical advantages that they would provide over fission triggered weapons: Pure fusion explosives can be made small enough to replace conventional munitions and also to fill the gap between conventional and current thermonuclear weapons. This advantage has diminished with time as nuclear weapons of smaller yields have been developed.1 Pure fusion weapons would produce no fission products. Most of the radioactivity produced would be in the form of short-lived activation products (notably argon-41). This would reduce political unacceptability and dangers to soldiers while maintaining the lethality of these weapons. Achieving both of these advantages simultaneously poses very great technical challenges. A smaller challenge may be to first develop hybrid fusion-fission devices that would not require a critical mass or the use of a fission explosion as a trigger. Jones and von Hippel have discussed such types of possible weapons in which the role of the primary and secondary in present-day nuclear weapons would be reversed. 2 A fusion primary would supply a sufficient number of neutrons to trigger a large number of fission reactions in uranium-238. Since neutrons from D-T reactions are highly energetic, they can fission U-238 (which cannot sustain a chain reaction, but which releases a large amount of energy when fissioned). Since each fission releases more than ten times as much energy as each fusion reaction, the fission secondary would serve to amplify the primary fusion explosion. This approach to nuclear weapons would negate one of the main reasons for seeking pure fusion weapons -- avoiding heavy radioactive fallout. But it could more easily achieve the first goal -- overcoming the "tyranny of critical mass"3 -- by using a fusion explosion as the primary part of the bomb. Once fusion-fission weapons are developed, the next "logical" step in the technical progression would be pure fusion weapons. Pure fusion also have a military "advantage" over conventional explosives in that the lethal radius is far larger than the range of explosive lethality alone (see below). B. Overall assessment of non-fission-triggered nuclear weapons There are two broad requirements for establishing the technological feasibility of non-fission-triggered nuclear weapons. First, non-fission heating and containment of a plasma must be achieved so as to generate sufficient fusion reactions to yield a net energy output. Second, a driver must be developed which both supplies the necessary power and can be made compact enough to be feasible as a weapon. We will evaluate each of these issues and then given an overall assessment of the technical prospects for non-fission-triggered nuclear weapons. 1. Ignition Existing ECF devices are being scaled up or modified to achieve ignition of a thermonuclear plasma. The explicit goal of a number of devices that are being built or designed is to achieve ignition. The National Ignition Facility, which is part of the US Science Based Stockpile Stewardship (SBSS) program, is among them. The Magnetized Target Fusion Program's status in relation to the SBSS program also aims to provide "scientifically exciting" research opportunities for scientists engaged in nuclear weapons design and testing. Its explicit goal is also to achieve ignition. If the National Ignition Facility achieves its stated design goals it should produce around 5-20 MJ of output energy during "high-gain" experiments. Its peak energy output would be approximately 45 MJ. The Laser Mégajoule project should have similar performance (though its peak energy output is estimated to be 60 MJ). 4 The energies and pressures of the NIF experiments can be compared to both previous fusion facilities and to nuclear weapons tests (Figure 11). As can be seen, many of the performance parameters approach nuclear weapons tests. For example, the energy density of NIF is very similar to nuclear weapons tests. What is different is that the total energy output is much lower in NIF. It should be noted that while NIF may approach weapons tests in a number of ways and thus could provide useful weapons design information, there are enough significant differences that applying information from NIF to existing nuclear weapons is highly problematic. Magnetized Target Fusion devices have already achieved neutron production of 1013 (10 trillion) neutrons per shot in the "warm" plasma even before implosion (separate implosion experiments have taken place to test the implosion of the liner). 5 A full test of the MTF/MAGO system (formation of a D-T plasma followed by implosion of the liner) is scheduled for 2000. 6 According to Jones and von Hippel's review of the MTF literature, this technology could achieve energy outputs equal to 1-10 gigajoules or 0.2-2 metric tons of high explosives. 7 The technical results achieved in the NOVA laser program at Livermore and the GEKKO XII program in Japan are comparable to those achieved in the MTF program, since they have also produced on the order of 1013 neutrons in one shot. However, in order for the ICF program to achieve ignition, a larger driver is needed, among other things. This is the objective in building NIF in the US, Laser Megajoule in France, as well as other proposed large ICF programs.
The MTF and NIF programs have somewhat different characteristics. Though the confinement time requirement for MTF is less stringent than for ICF, the physics of MTF is more complex and the MTF program only plans for a very limited number of shots. And the goal of MTF in its present configuration is ignition of the whole volume of fuel at once, which is inefficient and requires a large amount of driver energy. 8 In contrast, NIF will be able to fire one shot every four to eight hours and its aim is to achieve spark ignition -- that is, an ignition of the central core of the fuel pellet. The programs are complementary in some respects. Since NIF uses precise lasers which can be fired frequently, it can be used to develop pellet designs for various applications, including MTF. Similarly, the results of NIF experiments could also be used to create advances in pellets for X-ray technologies such as Sandia's wire-array z-pinch, which may be more suitable for pure fusion weapons. For instance, NOVA and NIF can be used to study the temporal shaping of x-ray pulses far better than the wire-array z-pinch. As another example, the MTF and wire array z-pinch are complementary, since the frequency of MTF experiments is very low. Hence, design of MTF devices driven by explosives could be helped greatly by experiments at other facilities such as PBFA-Z at Sandia because the MTF and wire-array z-pinch are the same in principle. They both use electromagnetic compression of a plasma by using a conductor carrying a high current. The pulsed power experiments at Sandia and ICF experiments with lasers are also complementary. For example, experimental results from the Saturn pulsed power facility (at Sandia) are being combined with experimental results from the NOVA laser at Livermore. The resulting information is similar to that expected to be generated by experiments to be generated at NIF. 9 Similarly, one can expect that results from NIF will be combined with results from Saturn, PBFA-Z, and X-1 to yield even more information about fusion ignition. For example, rather than seeing NIF and X-1 as competitors, they are considered complementary and research on NIF would aid in designing experiments for X-1. According to Donald Cook, director of Sandia's Pulsed Power Sciences Center, "Without the knowledge of target experiments from NIF, it would take considerably longer to achieve high yield on X-1, and the risk of failure would be greater." 10 Computers, such as those being developed for the Accelerated Strategic Computing Initiative (ASCI) would likely be used to achieve a high degree of coordination between various fusion programs. For instance, the data from NIF could be modeled using the software and hardware of ASCI. This could then enable design of targets that more closely match the requirements of the pulse shape from the x-ray machines. This kind of coordination between various kinds of initiatives and computer modeling has precedent in the design of thermonuclear weapons. The energy release from the primary (analogous to the driver in ICF and MTF) has a certain temporal and spatial profile. In the development of thermonuclear weapons, the design of the secondary must take this profile into account. The design of pure fusion weapons could proceed along similar lines, once the scientific feasibility of the concept has been established. 2. Drivers In addition to a properly designed target, a driver powerful enough to dump sufficient energy into a small fuel pellet to ignite the thermonuclear explosion is necessary. Many problems associated with the driver need to be solved to ignite a pure fusion explosion: The driver must deliver the energy to the fuel pellet uniformly to within a very narrow tolerance, so as to achieve a symmetrical explosion.
Of the drivers that are used in research that could be applied to pure fusion weapons, lasers and accelerators are clearly ill-suited, since they probably cannot be made small enough in the near future. Therefore, the main function of these technologies for the development of pure fusion weapons is to demonstrate that ignition is possible and to study and replicate the specific conditions under which ignition is achieved. Each shot in laser fusion or accelerator devices is relatively inexpensive once the machines are built and a commitment to operate them has been made. At the present stage, the potential drivers for pure fusion weapons are:
The use of chemicals alone is impractical for achieving ignition in ICF systems because of the slow speed of detonation and the difficulty of transferring the energy from the chemical explosive to the fuel pellet. However, various high-explosive-driven systems can be coupled with other electrical, magnetic or electromagnetic systems to create more suitable design approaches for pure fusion weapons. In all such systems, advanced materials manufacturing approaches as well as advanced materials could make the achievements more feasible -- for instance via the development of faster chemical explosives, smaller capacitors, and better ablators. Advanced technology using "engineered multilayers" could result in power electronic capacitors that are up to a hundred times more compact than those made with more conventional technology. 11 The most immediate prospects for miniaturization of the driver would appear to be the use of a combination of chemical explosives and miniaturized capacitors for generating high electrical currents. Drivers could also be made smaller by improving the efficiency of the driver-ablator system. The overall efficiency of the accelerator-driven driver-ablator system is at best only on the order of one percent in the case of accelerators. It is even lower with lasers. The low efficiency stems from both the low efficiency of energy conversion in the driver itself and the low efficiency of the coupling of the driver to the ablator. C. Overall technical prognosis for non-fission triggered nuclear weapons Since the breakeven point between driver energy output and fusion energy output has yet to be achieved, the scientific feasibility of pure fusion weapons has not yet been established, as we have discussed. However, advances in research in the last decade, and notably in the last few years, have brought the field to the point where the development of such weapons is, for the first time, a distinct possibility. The MTF apparatus is at present the most compact device available, if used with chemical explosives. A battery and a chemical explosive can be used as the sources of energy for the driver. A substantial neutron output has already been achieved. And no further basic conceptual breakthroughs appear to be necessary for the achievement of ignition. Jones and von Hippel have made comparisons of the lethal effects of a MTF device based upon current technologies with other weapons in order to evaluate the weapons potential of MTF. According to their calculations a system weighing 3 metric tons would have a total yield of 0.5-2.5 metric tons (with about 320 kg coming from actual high explosives). Assuming a one-ton TNT equivalent explosion, the blast effects would only come from about one-fifth of that yield. The rest of this energy would be in the neutrons. This is obviously of very little (if any) advantage over conventional high explosives if blast alone were the criteria. The blast effects could be improved by placing a layer of U-238 around the device which would fission due to the fast neutrons from the fusion reaction. Of course, this would also increase the weight of the device. Even such a crude fusion weapon would be militarily far more lethal than a conventional explosion because the neutrons increase the lethal radius of the weapons. They would deliver a lethal dose of radiation out to a radius of 100-500 meters depending on the presence of buildings. Table 3, taken from Jones and von Hippel compares the lethal effects of conventional high explosives, an MTF weapon and sarin. Their conclusion is that MTF would be comparable to chemical weapons in lethality. The advantage of MTF weapons could increase if researchers are able to achieve fusion in a "spark ignition" mode similar to ICF rather than the less efficient "volume ignition" mode which is the expected characteristic of MTF technology. Finally, it is crucial to note that the radius of small pure fusion weapons per unit of explosive power would be far greater than that of large nuclear weapons. 12 For instance, the destructive are per ton of TNT equivalent of the Hiroshima bomb was about 0.5 x 10-3 km2, which is a hundred times smaller than the lethal radius of a one ton TNT equivalent pure fusion bomb. 13
It is significant to note that these calculations are based on current technology and do not take into account future technological development and potential improvement in chemicals, batteries, or the magnetic flux compression generator itself. It also does not include the dramatic increases in efficiency, and hence yield, that could be achieved if magnetic of electrical compression could be used to generate "spark ignition" in a plasma. As discussed above, the driver energy required would then be greatly reduced for the same fusion energy output. Thus, the overall energy gain would be much higher, increasing the practicality of a pure fusion weapon. Moreover, it is remarkable that it appears that an existing apparatus could, in principle, be used to make a pure fusion explosive, though it should be kept in mind that ignition has not yet been achieved. The very possibility shows that we may be on the verge of a qualitatively different era in nuclear weapons. This is because there is generally a far smaller gap between the achievement of technical feasibility and a workable weapon than between initiation of scientific research and the establishment of technical feasibility. Recall that the first fission-driven thermonuclear explosion was not a weapon at all, since it was far too large to be deliverable. It required considerable design changes to achieve a deliverable weapon. Yet such changes were accomplished in less that one-and-a-half years. This is largely because data from the successful explosion enabled a critical revaluation of pre-explosion experiments and theory. Another technology that complements ICF and MTF programs is the wire array z-pinch developed at Sandia National Laboratory (see Chapter 3). It has already achieved an x-ray energy of 2 MJ and power of 290 trillion watts (terawatts), for a few nanoseconds. 14 This huge x-ray power could be focused on the ablating surface of a fuel pellet. Since this closely follows the Teller-Ulam approach to ignition of the secondary, the wire array z-pinch is considered an important tool for weapons research. Note that the x-ray energy already generated in the wire-array z-pinch is larger than the 1.8 MJ of laser energy planned for NIF several years into the next century. The next z-pinch facility desired by researchers is called the X-1; it aims for an x-ray output of about 16 MJ. 15 In sum, Magnetized Target Fusion, the ICF experiments in NIF and Laser Mégajoule, and x-ray generation and plasma compression experiments in the wire-array z-pinch at Sandia could together provide powerful ways in which pure fusion weapons or fusion-driven nuclear weapons could be designed. Experimental results from these programs could yield pure fusion weapons designs on a far shorter time-scale than would be possible with the MTF program alone. D. Fusion power and fusion weapons - comparative requirements Once the technical feasibility of pure fusion weapons is established, the weapons could be created in a variety of sizes. If a laboratory-scale D-T mixture is burned with an efficiency of thirty-three percent (a typical efficiency for planned ICF machines), then the explosive yield would be equivalent to about 20 kilograms of TNT16 A surface explosion of this size would create a crater about ten feet in diameter. 17 At the other end of the spectrum, huge megaton-size explosions can also be created from fusion reactions. This happens when thermonuclear weapons are detonated. Hence, in contrast to fission weapons, where even "small" explosions are very large - usually the equivalent of hundreds of tons of TNT -- pure fusion weapons could range in explosive power from small to huge. 18 ICF has also been proposed as the basis for possible commercial power production because explosions of tens or even hundreds of kilograms of TNT equivalent can be contained inside vessels. In an ICF scheme using five-milligram fuel pellets, about 5 explosions per second would be sufficient for a 1,000-megawatt (electric) power plant (about the size of a commercial nuclear reactor). However, achieving this rate of explosions in a practical ICF machine and extracting the energy contained in the neutrons from the reaction chamber poses severe technical challenges Most of the problems that need to be solved to produce pure fusion weapons are the same as those that face power production from inertial confinement fusion. As a result, though the initial impetus for ICF programs was weapons design (more specifically to help in the design of conventional two-stage thermonuclear weapons), they have, in time, come to have a dual justification -- one for the development of weapons and the other for the development of commercial power from thermonuclear reactions. Even today, producing pure fusion weapons (as opposed to conventional thermonuclear weapons with fission triggers) is not a stated goal of the program, so far as public information goes. However, the degree of difficulty of producing pure fusion weapons, while enormous, is lower in most respects than that confronting commercial power production from ICF. There are two broad sets of reasons that pure fusion weapons could be built before pure fusion power plants. The first involves economics, the second the technology needed for fusion power production. First, economics is not as much a central consideration in weapons development as it is with commercial power applications. For instance, the efficient use of tritium and deuterium, central to commercial power production, does not pose the same severe constraints when the objective is to develop pure fusion weapons. The reduced efficiency requirements for pure fusion weapons, while still enormously challenging, could probably be achieved sooner than the very high efficiency explosions needed for commercial ICF devices. Another economic consideration relates to fuel pellet design and cost. Commercial fuel pellets must be very cheap to be competitive - far cheaper than need be the case for fusion weapons. Secondly, there is no energy capture and conversion step needed for weapons. The arrangements to capture the neutron energy from the D-T reactions outside the reaction chamber, the generation of tritium fuel on a continuous (or near-continuous) basis, considerations of durability of machines under heavy bombardment of macroscopic explosions and of neutron radiation together constitute formidable obstacles to ICF fusion power development. Finally, there is a further loss of efficiency by a factor of two or three in conversion to electricity. Weapons design does not have to contend with any of these problems. The main technical issue that is more difficult with weapons than power is that weapons require compactness. Specifically, the miniaturization of the driver that is required for pure fusion weapons poses major challenges. Similar arguments can also be made for MTF research. In fact, as Jones and von Hippel have noted, the more efficient a laboratory device becomes, the greater the concern regarding the potential of the device to lead to pure fusion weapons. 19
| ||||||||||||||||||||||
Institute for Energy and Environmental Research
|
|