Nuclear weapons, to quote Sidney D. Drell, are "sophisticated but not complicated." That is, the working principles are straightforward, although the equipment needed to make a device function, and function reliably, is quite sophisticated and requires high-quality engineering to design and build. Although it is generally believed that a proliferator need not test a conservatively designed device at full yield to have confidence in it, some experimentation and testing along the way is necessary to demonstrate the behavior of the non-nuclear components including the firing set, detonators, and neutron generators. If there is not to be a full-yield nuclear test, then the non-nuclear experiments must be carried out with greater care and competence. One reason for believing that a full-yield nuclear test is unnecessary is that each of the six states known to have tested nuclear devices has achieved a nuclear detonation on the first try. The first nuclear weapon used in combat used an untested gun-assembled design, but a very simple and inefficient one. The first implosion device was tested on July 16, 1945, near Alamogordo, New Mexico, and an identical "physics package" (the portion of the weapon including fissile and fusion fuels plus high explosives) was swiftly incorporated into the bomb dropped on Nagasaki.
The term "nuclear testing" encompasses all experiments in which special nuclear material (or a simulant) is placed in contact with high explosives, which are then detonated, or with a propellant, which is ignited. This limitation deliberately excludes activities which are more scientific in nature and not intimately connected with the progression from fissile material and/or fusion fuel to a nuclear explosive device. This definition is far broader than that of the Comprehensive Test Ban Treaty (CTBT) of 1996, which prohibits only nuclear weapon test explosions and other nuclear explosions. Many states of concern for nuclear proliferation have subscribed to the CTBT, and may, therefore, find it difficult to conduct full-yield tests either underground or in the atmosphere. At the lowest end of the nuclear yield distribution from hydronuclear tests, some states might reckon that the knowledge gained from a small explosive release of nuclear energy would be worth the risk of getting caught. Generally, within the U.S. Government, the condition of prompt nuclear criticality distinguishes, under the CTBT, a prohibited test of an explosively assembled device from one which is allowed.
Nuclear Yield TestingFundamentally, test programs can be divided into two major categories: those for an HEU-fueled, gun-assembled device and those for an implosion device using either plutonium or HEU. The first Chinese test was of an HEU implosion device, Iraq intended to develop just such a weapon, and the South Africans conducted no nuclear tests of their gun-assembled devices. The general design of a gun-assembled device is straightforward and based on well-understood principles of artillery weapons; however, the technology for obtaining enriched uranium is complex. On the other hand, implosion-assembled devices using plutonium--which could be extracted simply using chemical techniques from reactor rods--are more difficult to manufacture. If a nation had an indigenous reactor industry, such extraction would be straightforward. The testing program for a gun-assembled device is moderately complex, but it is essential to realize that nothing nuclear need be tested to verify the probable operation of such a device--only its conventional components. The design of Little Boy, the bomb dropped on Hiroshima, had not been proof tested before the war shot. Implosion Devices
The testing program for a simple fission device using plutonium must be more extensive than that for a gun-assembled device using enriched uranium. For example, the constructor must know that his fissile "pit" will be uniformly compressed and that the compression will be rapid enough to minimize the chances for a pre-initiation "fizzle," that any neutron generator present will fire at the correct moment, and that compression is likely to be maintained long enough to result in significant nuclear yield.A proliferator hoping to demonstrate its technical prowess may elect to pursue an implosion device despite the availability of enriched uranium. Alternatively, it may choose implosion to achieve greater efficiency in the use of special material. It can be presumed that this type of proliferator will forego the development of thermonuclear weapons.
From 1945 through much of 1991, the United States detonated more than 1,200 nuclear devices with yields from a few pounds to about 15 megatons. Until the middle of 1963, most U.S. (and Soviet) tests took place in the atmosphere; some were con-ducted underground, a few were below the surface of the ocean, and roughly a dozen American shots took place at altitudes above 10 km. The largest test ever conducted, that of a 60-megaton device, was carried out in the Arctic by the USSR. Since the Limited Test Ban Treaty (LTBT) was signed in 1963, all U.S., UK, and Soviet nuclear detonations have been underground. The French and Chinese, while not parties to the LTBT, gradually moved their testing from the open atmosphere to subterranean sites-- in boreholes, mine shafts, and in drill holes beneath the ocean floor.
Atmospheric tests are easier to carry out --- although impossible to conceal --- and for technically less-sophisticated powers provide more information in a more direct manner than do underground explosions. A weapon detonated from a several hundred foot high tower or suspended from a tethered balloon permits photography of the evolution of the nuclear fireball and the cloud. The shock wave in air can be observed, and one can determine the effects of the weapon on real targets such as structures and vehicles.It appears likely that the drilling technology needed to emplace nuclear devices and instruments at the bottom of a deep borehole is the most difficult for a proliferator to acquire and use. Such boreholes are frequently a kilometer or more deep and 2 meters or more in diameter. The specialized drilling machinery required for such construction is not commonly available and exceeds what is found in the oil industry.
The development of the fireball and the propagation of a shock wave proceed quite differently when the device is tightly tamped at the bottom of a borehole than when it is detonated in free air. However, when the borehole or mine shaft have been properly stemmed, underground experiments have the advantage of not releasing significant amounts of radioactive debris. It is also simpler to place large masses of experimental apparatus close to an underground shot than to locate the same hardware next to a balloon gondola or on the platform of a slender tower, either of which has a limited carrying capacity. In any event, very few atmospheric tests have been carried out during the last three decades, and even the French and Chinese abandoned their atmospheric test programs.
Only with a large collection of data derived from yield tests of different types of devices can a weapons designer be confident that he understands the behavior of different possible designs within what is termed the nuclear weapons "design space," and only then can he be confident that the computer programs used to predict device performance deliver reliable results. This may be the strongest motivation for a proliferator to test at full yield. However, even a series of full-yield tests may not provide all of the information needed for weapons design.Most nuclear weapon states have constructed underground testing facilities similar to the U.S. Nevada Test Site. That is, weapons development and proof tests are usually carried out in vertical shafts stemmed to prevent the escape of radioactive debris. Radioactive debris from an atmospheric test or from an underground shot which vents can be analyzed by other nations. Much information about the design and performance of the test device can be inferred from the debris.
Power and signal cables for the device are routed up the shaft and fanned out to several instrumentation trailers outside the probable cratering zone. Nuclear weapons effects tests are primarily carried out in horizontal mine shafts sealed to prevent the escape of debris; instrumentation cables are connected to the surface through a vertical bore hole. In both cases, the tests are characterized by the large amount of electronic instrumentation used to study the details of the functioning of the implosion assembly and of the nuclear phases of the explosion. A beginning nuclear power opting for simpler weapons may well choose not to employ sophisticated diagnostic instrumentation, selecting instead to determine the approximate yield with seismographs.The most accurate measurement of yield is through the radio-chemistry studies of device debris --- the radioactive isotopes produced in the detonation. No electronics are used to gather the data for such analyses; it is only necessary to drill back into the device chamber and to extract samples for lab examination. A faster but less accurate yield determination can be done using seismographs to measure ground motion, but such a test would not collect a large quantity of data usually considered desirable by U.S. weapon designers and testers.
In a hydronuclear test, fissile material is imploded, but a supercritical mass is not maintained for a long enough time to permit the device to deliver "full" nuclear yield. Depending upon the conditions of the test, nuclear energy releases may range from the unmeasurably small (milligrams or less) to kilograms or even metric tons of TNT equivalent yield.
Hydronuclear experiments, as distinguished from hydrodynamic ones, use actual fissile material assembled to form a supercritical mass in which a chain reaction be-gins. Normally, hydronuclear experiments are designed to use nuclear devices modified in one of several ways, including substituting inert material or less-fissile material for some of the HEU or plutonium in the pit, so that very little nuclear energy release occurs. Yields in experiments described as "hydronuclear" by various countries have ranged from much less than 1 kg TNT equivalent to many tons.
In a hydrodynamic test, inert material (e.g., 238U or a simulant for plutonium) is imploded to determine how well the high-explosive system functions. The testing program for an unboosted implosion device primarily ensures that the hydrodynamic behavior of the implosion (particularly of a hollow pit) is correct. The simplest way to do hydrodynamic testing is to implode inert pits made of a simulant for fissile material (e.g., natural uranium instead of HEU) while using any of several "old fashioned" means to observe the behavior of the heavy metal. One such technique is to use a pin-dome, essentially nothing more than a precisely machined insulating "champagne cork" with a large number of protruding radial pins of different distances placed at the center of the implosion region.
Pin dome experiments are probably the easiest hydrodynamic diagnostics available. However, backlighting the pit with a flash x-ray or neutron source to obtain an actual picture of the imploding material is also a possibility. Generally, the flash x-ray source needed has to have very high peak power available in a single pulse, and the timing and firing of the source in concert with the implosion of the device requires very sophisticated system design. Backlighting the imploding system with a neutron source is a bit more straightforward, but requires very sophisticated neutron optics and imaging capability, which could be difficult to obtain. Iraq used flash x-ray diagnostics.
The Radio Lanthanum (RaLa) method, which does permit time-dependent measurements of the symmetry of an implosion, should be mentioned because of its conceptual simplicity. RaLa was used extensively during the Manhattan Project, but has probably not been employed very often since then. An intensely radioactive sample of the element lanthanum was prepared in an accelerator or reactor and then quickly inserted into the center of the implosion test device. Highly collimated Geiger-Mueller counters observed the behavior of the material as it imploded. The RaLa technique is inherently fairly crude in its ability to detect asymmetries and environmentally unappealing because the radioactive material is scattered about the test stand. However, the isotopes have half lives of only a few hours to a few days, so the residual radioactivity decreases significantly in a week or so.
Nuclear Weapons Effects TestingSome nuclear weapons effects (NWE) can be modeled mathematically using powerful computers; others, and in particular the combination of several effects, are beyond valid analytic or numerical assessment. The only way to know if friendly systems or target assets will endure a given nuclear attack may be to expose representative equipment to real nuclear explosions or to construct complex simulators which reproduce a part of the spectrum of NWE. Until the conclusion of the Limited Test Ban Treaty (LTBT) in 1963, the United States conducted atmospheric tests of nuclear weapons, and it was relatively simple to include effects testing in the experiment. By signing the 1963 accord, the United States, the UK, and the Former Soviet Union agreed to discontinue atmospheric testing, testing in outer space, and testing under water. The only environment in which nuclear devices could be detonated was underground in circumstances where radioactive debris did not drift beyond national boundaries.
In the years between 1963 and 1992 the States Parties to the LTBT conducted underground tests to study NWE. As a result of congressional action the United States unilaterally entered a testing moratorium, which was made permanent with the signing of the Comprehensive Test Ban Treaty (CTBT) in 1996. Because it is no longer considered acceptable for the United States to conduct any nuclear explosions for any reason, future U.S. assessments of the vulnerability of its systems or of potentially hostile systems will have to rely upon the use of simulation and analysis validated by comparison with the results from almost 50 years of testing.
Combinations of nuclear weapons effects pose particularly difficult simulation problems. The thermal pulse can weaken or ignite a target, permitting the blast wave to be more effective than against a "cold" object. X-ray radiation can damage elec- tronics and protective systems, making the target more vulnerable to neutrons. EMP and transient radiation effects in electronics (TREE) can operate synergistically. Thermal effects could conceivably damage some components designed to harden a system against EMP. Low-energy x-rays absorbed by a target in space can heat surface material to the vaporization point, causing it to explode away from the system, producing shock effects within the target. The effects produced and the ranges at which they are effective depend upon the yield of the nuclear weapon and the height of burst (HOB) and may depend upon the design of the device itself.Theoretical predictions of NWE based on computer codes and algorithms that have not been compared with experiments may not be accurate, and the details of such experiments are not generally available. Those codes and algorithms which have been validated by experiment usually contain adjustable parameters and are much more reliable predictors of NWE. Such codes are termed "substantiated." Physical simulation provides more confidence in predicting NWE because it does not rely upon the mathematical approximations of codes and algorithms but uses physical phenomena closely related to those produced by a nuclear detonation to test the behavior of real systems. But physical simulation remains "second best" compared to testing against a real nuclear detonation.
Underground testing (UGT) can provide much insight into weapon design, radiation effects (gammas, neutrons, x-rays) on military systems, selected aspects of shock and blast, thermal effects, and source region EMP (SREMP). Countries with limited defense budgets are less likely than the major nuclear powers to have had exhaustive underground testing programs.
An understanding of TREE and System-Generated Electromagnetic Pulse (SGEMP) is of critical importance in designing and building equipment that can survive a nuclear attack. It is not clear, however, that a nation having limited financial and technical resources could develop unique radiation-hardened devices and/or systems. These countries could, however, test a few critical subsystems or systems in an established foreign simulation facility. Although there are certain aspects of TREE and SGEMP technology that are of general scientific interest, for nations which have interests in the acquisition of nuclear weapons, the desire to evaluate and test systems at SGEMP and TREE dose rate levels typical of nuclear weapons is a useful indicator that they plan on nuclear combat, whether as a user or as a victim of the weapon. While TREE and SGEMP may indeed be effective, a nuclear planner without the benefit of extensive simulation and substantiated codes will probably rely on the gross NWE such as blast, shock, and thermal radiation.In the absence of nuclear testing, simulation equipment, numerical simulation, and theoretical analysis of NWE are the only means states can verify how NWE will affect their own forces and those of their opponents in a nuclear environment. NWE simulation, as well as survivability and hardening programs, have both offensive and defensive aspects, and may be desired by both nuclear possessor states and those with neither nuclear weapons nor plans to build them.
Most of the relevant equipment and specialized software has been developed in parallel by many countries including Russia, China, the UK, and France, as well as Japan, Germany, Switzerland, Sweden, Canada, and members of the former Warsaw Treaty Organization. Although the simulation, survivability, and hardening equipment available from non-Western countries is inferior to that produced in the West ("years behind" in the case of HEMP simulation), it may be good enough to permit a nuclear aspirant to understand how to make its own equipment more survivable than otherwise. The most advanced capabilities usually only are necessary when one is trying to design equipment to be the lightest, most effective, and most efficient; when one backs away from the edge of the envelope, less-detailed analysis and testing may suffice. After all, the NATO allies operated acceptably survivable equipment decades ago.
Underground Nuclear Weapons Effects TestsUnderground nuclear weapons effects tests (UGWETs) provide nuclear environments for demonstrating the hardness and surviv-ability of military equipment and materials as well as for studying basic nuclear effects phenomenology. Full-yield nuclear tests are the only way to produce all relevant nuclear weapon effects simultaneously. Underground nuclear weapons effects tests can provide insight into weapon performance, nuclear radiation effects, shock and blast, thermal effects, and source region EMP (SREMP). Even when it was allowed, underground testing was a very expensive way to garner the needed information. It was used by countries with significant economic bases and which were also committed to the development of nuclear offensive and defensive capabilities.
The UGWET-specific technologies include horizontal emplacement of the de-vice, the provision of evacuated horizontal line-of-sight (HLOS) tubes for viewing the detonation, and mechanical closures to prevent debris from traveling through the HLOS tube to the experiment station that measures the radiation and shock environment and the response of systems. Also included are scattering station design and the computer codes necessary to understand the results of the experiments. Technologies to contain the release of radiation are only covered to the extent that they differ from those used in nuclear weapon development tests.
For effects testing, horizontal emplacement tests (HET) are preferred over vertical emplacement tests because the emplacement of device and test equipment is simplified. Horizontal tunnels provide greater experiment flexibility and access. Vertical shaft tests are less expensive but only provide limited exposure area because of the risk associated with containment when the crater is formed. The need to excavate large cavities for the placing of "test samples" and the construction of appropriate environments for those samples (for example, a vacuum for reentry bodies) drives the conductor of HLOS tests to seek suitable terrain such as a mesa or mountainside. Effects tests could also be conducted inside a deep mine.
HETs can incorporate large cavities so that shock and SREMP from a low-yield device actually have space to develop to the point where they are representative of similar effects in the open air from a large-yield weapon. The minimum burial depth is:
D = 400 Y 1/3 feet,and the radius of the cavity formed by the detonation is: R = 55 Y 1/3 feet,where linear dimensions are measured in feet and yield in kilotons. The object of an HET is often to allow nuclear radiation to reach the test object while preventing it from being destroyed by the other effects. Indeed, scientists expect to be able to recover the test instrumentation. Such a test requires redundant containment vessels: the first around the device, a second around all of the experiment to protect the tunnel system if the inner vessel fails and the experimental equipment is lost, and a third to ensure that no radiation escapes into the atmosphere even if the experimental equipment is lost and the tunnel system contaminated.
The HET-HLOS configuration is most often used for radiation effects tests, but the HLOS configuration must withstand the blast and shock waves produced by the device. The HLOS pipe is tapered from about 6 inches in diameter at the "zero room" (the device emplacement cavity) to about 30 feet in diameter at the experimental area 1,500 to 1,800 feet away and provides a clear line of sight to the device for those test subjects which need to see direct radiation.Not all experiments require "direct" nuclear radiation; many are suitable for use with a scattered (lower intensity) beam produced in a scatter station--typically made with appropriate nuclear and atomic properties to deflect the correct wavelength and intensity of radiation. The design of these scatter stations requires both technical skill and experience so that the scattered radiation is properly tailored for its intended use. An incorrectly designed station could mean that the test object is exposed to incorrect radiation types or intensities, which could significantly reduce the value of the test.
Complete containment of radioactive debris is probably essential if a nation wishes to conduct a clandestine nuclear test. In any underground nuclear weapons effects test (UGWET), fast-acting mechanical closures to prevent debris from reaching the test objects are unique and critical equipment. A number of techniques are used in parallel to ensure that the HLOS pipe is closed before nuclear debris reaches the experiment. X- and gamma-rays travel at the speed of light, and electrons (beta particles) and neutrons are not much slower. The debris, however, moves much more slowly, at hydrodynamic velocities. [A "modified auxiliary closure" (MAC) or, when lower-yield weapons are used, a "fast acting closure" (FAC), positioned close to the device location--the working point--is able to shut the pipe in about 1 ms and to withstand pressures of about 30,000 psi.] A gas seal auxiliary closure (GSAC) farther along the HLOS pipe can close in less than 30 ms, and the tunnel and pipe seal (TAPS) will shut the pipe off in 300-700 ms. The TAPS is considerably farther from the working point than the FAC and therefore (a) has more time to function and (b) must close a larger aperture due to the taper of the HLOS pipe. These closure technologies are likely to require significant experience to develop to the point of reliable operation.
Other instrumentation to measure device performance, delivered shock, thermal pulse, electromagnetic pulse, and radiation is essentially similar to that used in a device development test.
Emplacement canisters, fast-acting closures for HLOS tunnels, and containment technology are the keys to preventing the release of radioactive debris into the atmosphere, allowing UGWET tests to be conducted without their being detected off-site. Mechanical closure designs and materials unique to underground tests in general and UGWET in particular include mechanical and cable gas-flow blocking designs and techniques that operate up to a pressure difference of 1,000 psi for up to an hour and specialized explosive and/or mechanically driven devices capable of isolating portions of the HLOS pipe during or within the first 100 ms after exposure to radiation.Because the experimental area is often quite large and is at a considerable distance from the working point, the vacuum systems needed to evacuate air from them to simulate a space environment are unusual. Required are specially designed diffusion or cryogenic pumps capable of maintaining a pressure much less than 10 -3 Torr over a pipe system as long as 1,800 feet and varying in diameter from as small as 1 inch to as large as 30 feet. The crystals used to determine the energy spectrum of the radiation are unusual as well, and must be specially designed and fabricated to measure x-ray fluences at levels >0.1 cal/cm 2 in a time <50 ns and to operate in the UGT environment.
Some foreign vendors can manufacture digitizers, measurement systems, and fiber-optic equipment comparable to those used in U.S. UGWET. France manufactures digitizing oscilloscopes; Japan, South Korea, and Taiwan manufacture the electronic components for measurement and recording systems; and Germany manufactures cryogenic vacuum pumps of the large size required for HLOS events.
Blast and Shock SimulationIn the absence of atmospheric and underground nuclear testing to determine the survivability of structures, means must be found to simulate the phenomena associated with a nuclear explosion. For blast and shock this can be done either in a large-scale, open-air test employing chemical explosives or in a specially designed test facility which can also produce thermal fluxes comparable to those from a nuclear weapon. The air blast from a nuclear explosion is, however, different from that produced by conventional explosives. Because of the intense thermal pulse, the surface and near-surface air mass surrounding ground zero is heated rapidly. Within this heated region the blast wave travels more rapidly than it does in the cooler air above. As a result, blast waves reflected from the ground travel outwards and merge with the direct blast wave from the explosion. This produces a nearly vertical shock front called the Mach stem, which is more intense than that from the direct blast. To simulate the Mach stem with tests using high explosives, scientists employed helium-filled bags at ground level surrounding the high explosives used in the test. Because such tests can only be scaled and do not replicate the actual effects of a nuclear explosion, only scale models of test objects could normally be used.
More recently, U.S. attention has focused on a higher pressure regime than can be attained in open-air testing and on the construction of large simulators capable of re-producing simultaneously the blast and the thermal pulse from a nuclear detonation. These simulators typically employ a fuel-oxygen mixture, for example, liquid oxygen and finely powdered aluminum, and consist of long semicircular tubes. These simulators can even approximate the effects of soil type on blast wave propagation as well as the entraining of dust in the blast wave.The actual combination of overpressure, dynamic pressure, lift, and diffraction effects on a target is exceedingly difficult to model analytically or to simulate numerically, particularly without actual data. Military interest in the effects of dynamic loading on systems is in the survivability of tracked and wheeled vehicles, towed vehicles, C 3 shelters, etc., in the pressure regime characteristic of nuclear weapons. Civilian interest is in the survivability of similar systems and structures subjected to storm winds. The two are not completely distinct interests because the dynamic pressure from strong hurricanes may be comparable to that from nuclear blasts. Military interest also focuses on shock loading, a dynamic process which differs from the nearly steady-state effects of storm winds. As a rule of thumb, a 30 kPa pressure threshold corresponding to a 60 m/s particle velocity in the shock, or a drag force equivalent to that produced by about 210 km/hr (130 mph) steady winds, distinguishes the military and civilian applications. A frequently used design objective for civil structures is survivability in 190 km/hr (120 mph) winds. Technologies for simulation include not only the ability to produce strong shocks and air blasts but also those used to measure shock wave values, dynamic pressure in a dusty environment, and deflections or other motions of the test structure. Dust-loaded shock tubes are unique to NWE testing. Similarly, combining both blast and thermal pulse would be unique to the nuclear situation. Explosives which are diluted or mixed with inert materials such as dilute explosive tiles produce more uniform detonations that more closely resemble a nuclear detonation; such explosives would also be critical to NWE testing.
Thermal SimulationMixing and ignition facilities with surface emittance rates on the order of 150 cal/cm 2-s at blackbody temperatures of ³ 3,000 K are critical to some simulators. Such mixer facilities should mix fuel and oxidizer before ignition to avoid the production of smokes and particulate clouds. Instrumentation designed to function at flux levels above about 150 cal/cm 2 -s is specialized to the nuclear simulation role; this intense radiation environment can easily melt all known materials over the duration of a full thermal pulse. These conditions are not found in any commercial applications. Other processes and technologies such as plasma discharges with arc diameters >1.0 cm and arc lengths >10 cm for current greater than 1,000 Å and more than 300 kW input power are unique to nuclear simulation and have no commercial appli-cations. Software is to be validated against nuclear detonations or simulations and intended to model the characteristics of the fireball as functions of the characteristics of the nuclear source, burst environment, and atmospheric conditions.
The new U.S. Large Blast/Thermal Simulator (LBTS) is the most advanced facility of its type in the West, having a larger operating envelope (blast) than the comparable French instrument plus the capability to perform simultaneous blast and thermal testing, also a capability lacked by the French. The United States and France lead in full-scale, thermal pulse simulation technology. Large-area, chemically driven, thermal-radiation simulators were developed in the United States but have been sold to France, the UK, and Germany. The United States operates flash and continuous-lamp facilities and uses solar furnaces on small targets. France and Germany have made incremental improvements to the simulators purchased from the United States. Russia and some Eastern European countries have thermal simulators comparable to those of the United States and other NATO nations.
Radiation, as commonly used in the nuclear weapons arena, applies to neutrons, gamma rays, and x-rays alike. It can also include high-energy beta particles (elec-trons). All of these types of radiation show corpuscular behavior when interacting with matter--the high-energy photons because of their extremely short wavelength. Describing these interactions quantitatively requires the full machinery of relativistic quantum mechanics including the computation of the relevant Feynman diagrams. The particle energies involved range from the upper energy limit of the ultraviolet band, 0.124 keV, to the MeV and tens of MeV associated with the gamma rays and neutrons emitted from a fissioning or fusioning nucleus.The distinction between x-rays and gamma rays is not fundamentally based on photon energy. Normally, one speaks of gamma rays as having energies between 10 keV and 10 MeV and thinks of even hard x-rays as having lower energies. In fact, the difference between the two phenomena lies in their origin: gamma rays are produced in nuclear reactions while x-rays are an atomic phenomenon produced by electron transitions between discrete atomic levels or by blackbody (thermal) radiation from a heated object. A reasonable upper bound for "x-ray energy" in discussing nuclear phenomenology would be a few hundred keV, associated with the initial stages of fireball formation.
Because actual nuclear tests can no longer be performed, and because above-ground explosions have been prohibited since 1963, the only ways to determine the results of attacks utilize simulators, theoretical models, and the data from earlier U.S. nuclear tests. The integrated use of this information in computer models which can predict the HEMP environment as a function of weapon parameters and explosion geometry is a critical technology requiring protection. In contrast, basic theoretical models lacking actual test results should not be controlled.
Theoretical models of HEMP coupling to generic systems such as cables and antennas are of general scientific interest. Codes associated with the generic coupling of HEMP to systems and which do not reveal specific features of military systems and their responses, performance, and vulnerabilities to HEMP need not be controlled. These codes are similar to those used in electromagnetic compatibility and electro-magnetic interference and the study of lightning. Interest in the synergism between lightning and HEMP will continue.The United States has been the world leader in HEMP technology since the first articles on the subject appeared in the early 1960's. These scientific papers appeared in the open literature, which allowed the Soviet Union to become active in the field. The general consensus is that Soviet (now Russian) capabilities lag years behind those of the United States. Nonetheless, Soviet interest in pulsed-power, which began under A.D. Sakharov, should call attention to the possibility that some of the Soviet HEMP program was very closely held. HEMP capabilities have been acquired by the European nations, including Sweden and Switzerland. Many of these countries have developed active programs that include the use of simulators operating nearly at the threat level. Papers presented at recent unclassified conferences by participants from the countries of the former Warsaw Pact indicate that they lag significantly behind the West in both simulation and theoretical understanding. Several foreign vendors produce equipment comparable to that available from U.S. sources. France manufactures pulse generators, field sensors, fiber-optic links, transient digitizers, and measurement systems; England manufactures 1-GHz band-width fiber-optic links used mainly in HEMP and conducts high-power microwave research. Switzerland and Israel have also developed test/simulation equipment of high quality.
The upper limit to the frequency of the electromagnetic radiation attributed to HEMP is in the range of a few GHz. Thus, the interactions of the HEMP pulse with systems can be computed using classical electromagnetic theory without the need to include quantum effects. Off-the-shelf equipment suffices for the simulation of HEMP in small volumes. The peak electric field is about 50 kV/m, with a pulse width of several nanoseconds. However, producing equivalent fields over an entire military system such as a tank requires a very large radiating system with feed-point driving voltages in the megavolt range. The combination of antenna feed-point voltage and nanosecond rise time is what gives rise to the connection between HEMP pulsed-power technology and the technology needed to produce appropriate gamma- and x-rays. The production of pulses of neutrons corresponding to those generated by a nuclear weapon is primarily of interest for simulating TREE.
Flash x-ray (FXR) techniques are used to produce hard and soft x-rays. Typically, a high-energy electron beam is dumped onto a target to produce bremsstrahlung ("breaking radiation") photons over a broad range of energies up to the kinetic energy of the incident particles. Calculating the actual spectrum produced in a given target is difficult because thick targets, in which the electrons may interact several times, are required to obtain the desired intensities. This, in turn, raises the importance of nonlinear terms. Ideally, an FXR device should produce the same photon spectrum distributed identically over time as the spectrum from a nuclear device. This is not possible at the present time, but existing simulators provide useful approximations.
Specific technologies used to provide the power pulse include the Z-pinch; Blumlein or coaxial cable pulse-forming and transmission lines; large banks of very high-quality, low-loss capacitors; fast opening and closing gas and liquid switches with very low resistance in the closed state; Marx generators to produce the actual high-voltage pulse, and even Van de Graaff electrostatic generators with high current (for the class of accelerator) output. The switches used are unusual and have few other uses. One, for example, must conduct with a low resistance over a period of 0.4 to 1.0 microsecond, but must open to a high resistance state in times of the order of 10 ns.
Pulsed-power generating and conditioning systems and their associated loads (e.g., vacuum diodes) which convert the pulsed system's electrical output pulse to a photon or particle beam are valuable tools to study the hardness and survivability of critical military systems. The required fidelity of the simulation increases as the size of tested hardware increases because it is important to maintain the correct conditions over the aggregate of components which must function together. Some aspects of systems used in simulators are unclassified, and some border on the classified world. Some devices which may be used to simulate nuclear effects (e.g., the National Ignition Facility to be built at Livermore, or the Particle Beam Fusion Accelerator operating at Sandia National Lab) are also research tools for the broader scientific community.
Of particular importance are NWE simulators that can produce pulses with peak power greater than 25 TW from sources with impedance <0.1 ohm and having vacuum power flow and conditioning that can couple to a radiating load having a circular area less than 500 cm 2. These performance levels exceed the publicly available figures for the SATURN and HERMES III accelerators at Sandia National Laboratory.Russia has demonstrated strong NWE simulation capabilities, comparable to those of the United States. The UK and France have extensive programs, but less ambitious than Russia's. China has an NWE simulation program, but little is known about its capabilities. Germany has always been a leader in pulsed-power conditioning for basic research applications. Pulsed-power conditioning has been developed in Sweden, primarily to support kinetic energy and particle beam weapons research; in Switzerland, to investigate pro-tection against EMP; and in Israel, primarily for basic research at the Weizmann Institute of Science and for kinetic-energy weapons research at Israel's SOREQ Nuclear Research Center. Germany and Japan use similar technology primarily in support of light ion beams for inertial confinement fusion.
For HEMP simulation, the principal advanced technologies developed in the United States for risetimes less than 2 ns are multiple channel gas switches and multistage circuits in which the last stage charges very rapidly to increase the breakdown field of the output switch and decrease its inductance. The existence of triggered multichannel switches and the use of multistage circuits has been reported widely, but not in the context of EMP simulations. Countries with substantial pulsed-power capabilities (e.g., the UK, France, Russia, and Japan) could easily develop EMP simulators using such technologies.
Every stage of nuclear weapon development, from material production to deployment, can generate signatures that provide some indication of a weapon program's existence or status, although only a few of them point fairly un-ambiguously to a nuclear weapon program.
Repeated high explosive (HE) tests are generally required before a workable implosion-type nuclear weapon can be designed. Explosive tests to study either the HE alone or its ability to propel metal objects would usually require electronic or optical instrumentation. Some indicators of high-explosive testing activity are the following:
Test-firing of HE-metal systems containing uranium would be indicated by the following:
- expansion of facilities or personnel at or near an existing ordnance plant;
- purchase or production of explosives more energetic than pure TNT, such as RDX, HMX, or PETN, any of which could be mixed with TNT;
- equipment for compacting or melting and casting HE, perhaps modified from what would be used at a standard ammunition loading plant;
- alternatively, for different types of explosives, isostatic or hydrostatic presses, weighing many tons and likely remotely controlled (some antitank shaped-charges are also made using such presses);
- precision, possibly template or computer-numerically controlled, two-axis machining facilities for HE, especially if suited for machining curved contours and surrounded by blast-protection shielding;
- waste and scrap from the above operations, possibly including effluent waste-water systems involving filters or catch basins;
- pronounced red coloration in waste water caused by dissolved TNT;
- solid scrap periodically destroyed by burning or detonation; and
- instrumented firing stations and control bunkers for HE or HE-metal tests using charges weighing up to hundreds of pounds.
Since highly dense (but nonfissile) uranium-238 is widely used in certain types of antitank weapon, these indicators could also stem from advanced nonnuclear munition programs. Therefore, most or all of these could be associated with conventional munitions production and do not give unambiguous evidence of nuclear weapon development. However, spherically symmetric implosions would be more likely connected with a nuclear program.
- bright streamers radiating from the test (caused by burning fragments of uranium) visible to the eye;
- local debris or dust that contained uranium; and
- nearby fire-extinguishing equipment, portable radiation monitoring equipment, or permanent air-sampling radiation-monitors.
Gun-type weapons generally require highly enriched uranium surrounded by neutron-reflecting material such as natural uranium, tungsten alloy, or beryllium metal or oxide (ceramic). A development program might use hundreds of pounds of beryllium or thousands of pounds of uranium or tungsten for the neutron reflector alone. Unusually high importation of some of these items by certain countries might suggest weapon-development activity. In addition, ground cover at the detonation test-site may be cleared in only one direction, since the debris from tests-and especially burning uranium streamers if natural uranium were used as a mockup for HEU-would be concentrated in a cone coaxial with the direction of projectile firing (however, a test program for nonfissioning shaped charges or kinetic-energy rounds could also have such a configuration); special fast-acting very-high-pressure gauges might be used to record the pressures in the gun breech; and distinct acoustic features might be observable.Observers who had access to suspicious laboratories might detect the following signatures:
- Criticality tests --- weapon designers using near-critical fissile assemblies may wish to measure criticality with closed-circuit television and neutron counters in remotely operated (possibly underground) experiments. However, experiments can also be performed at the bench-top level, not needing elaborate equipment, and much of the relevant data is already available in the open literature. Moreover, similar facilities are also used for agricultural and biological neutron-irradiation research. Any kind of criticality accident at a suspect site, however, would be a strong indicator of weapon-design activity, since other applications would be ununlikely to work with near-critical assemblies.
- Neutron background measurements --- for gun-type devices, neutron-flux measurements would be required to assure that background neutron counts were sufficiently low. Such measurements might be indicated by a room containing neutron detectors that was shielded from external sources of neutrons, for example with water- or polyethylene-filled walls. Such facilities might also be used to test neutron initiators.
- Development of neutron initiators -- neutron initiators produce a pulse of neutrons to initiate the nuclear chain reaction at the optimum moment. They use either alpha-particle-emitting radioactive substances or small particle accelerators con-taining radioactive tritium gas. Therefore, import or production of alpha-emitting materials, tritium, or the special facilities to handle them (similar to those used for spent-fuel reprocessing) could indicate weapon development. However, small accelerator-based neutron sources are produced commercially for oil-well logging and laboratory use, so that they do not necessarily indicate a weapon program.
- Special tests --- Since neutron initiation is so important to the proper detonation of a nuclear device, tests involving actual HE with very small (sub-critical) amounts of nuclear material would likely be carried out as well. These might be conducted in shallow underground chambers designed for neutron shielding. (A series of such hydronuclear experiments was conducted by the United States during the testing moratorium of 1958-61.) Some of the surface equipment associated with these tests might also be telling.
Regardless of whether a country chooses to test a nuclear device at full yield or at reduced yields, a suitable underground site would be highly desirable. Since underground tests can be contained quite effectively when carried out properly, test-preparation activities would often be more observable than would atmospheric releases from tests themselves. Drilling rigs, sections of one-meter-diameter or larger pipe, mining operations, or road construction in new remote locations could all indicate such preparations and could probably be observed by reconnaissance satellites. (Determining that such activities actually do pertain to nuclear testing, however, may prove more difficult.) Large drilling equipment, cranes, heavy electrical cables, and roads could all provide visual indicators of such a test site. Electronic data-acquisition systems, which are widely available around the world, would require extensive cabling systems suitable for transmitting diagnostic signals might also be visible.