early days of operation to supervise the accountability for uranium materials within the plant. Adequate controls were secured only with the whole-hearted cooperation of the management and workers, when they were convinced that it was their patriotic duty to help reduce losses. I am certain that, had my group been the staff of an international inspectorate in a plant operated by a nation intent on diversion, we could not have stopped the diversion from taking place. The only effective way to reduce the possibility of such diversions is for the plant to be managed by an international control agency. The possibility of adequately controlling parts of a process by inspection has been suggested. In an electromagnetic process, there is no such possibility, since from the first step in the process until the preparation of the final product there are no logical points of division between inspection and management. Valuable material is always available in the process areas and in a form tempting to those bent on diversion. The final product from an electromagnetic plant can be made suitable for use in atomic weapons with little additional work. Therefore, this operation is a dangerous one and can be controlled adequately only if the operation as a whole is carried out under complete management by the international control agency. The electromagnetic process is one of those methods for isotope separation that is susceptible to major improvements. A new trick of operation might result in great increases in production. An inspectorate might fail to detect such a trick until dangerous amounts of product had been diverted to a clandestine bomb plant. The international control agency in operating electromagnetic plants could keep ahead in this field and thus be in a position to prevent such diversions. Clandestine Plants The detection of clandestine operation of an electromagnetic isotope separation plant should be considered. The most obvious method of detecting such an operation would be to obtain information on plants using large amounts of electric power. Large power consumption is a characteristic of the electromagnetic process. It is possible that, to avoid detection, many small plants would be constructed so that their individual electrical demands could be concealed. method of operation would increase the chances of detection because of the large number of units involved. The discovery of any one of these would call for investigation by the international authority and would probably result in the discovery of the remaining units. Electromagnetic plants use large quantities of water and would probably be located near lakes or streams. This would also help in locating clandestine operations. Seizure Seizure of an electromagnetic plant might prove less profitable than seizure of other types of plants in that there might be more time for destruction of the many separate buildings, but the size of the inprocess inventory in a large electromagnetic plant might prove tempting to a nation preparing for an atomic war. In-process material and reserve stocks should be kept as low as possible. 5. Control of the Gaseous Diffusion Isotope Separation Plant by George T. Felbeck Dr. Felbeck is vice president of Carbide and Carbon Chem- Introduction This article is on the subject of the control of a plant for the concentration of uranium-235 by the gaseous diffusion method. Since the problem is similar to that for the electromagnetic method, much that is said here will be a repetition of what appears in the previous article by Mr. Kelley. It is assumed that the method of separating isotopes by gaseous diffusion is already familiar in a general way, since it has been described in the Smyth Report and elsewhere. However, at the risk of repetition, a few of its fundamental principles which are pertinent to this discussion should be mentioned. Unlike other projects of the Manhattan District, the gaseous diffusion method does not involve the use of any of the ideas of modern nuclear physics beyond the concept of the existence of isotopes. The basic principles involve only the old-fashioned classical billiard-ball physics which is encompassed by the kinetic theory of gases. All of the desired results can be obtained by thinking of molecules as little round balls of various weights and forgetting all about neutrons, nuclei, electron clouds, radioactivity, and all the other ideas that concern a modern physicist. The fundamentals of kinetic theory were evolved by Rumford, Maxwell, Clausius, Kelvin, and Boltzmann in the period of 1850 to 1880. The application of the principles of kinetic theory to the separation of gaseous mixtures was made by Wussow, Hertz, and Aston in the period of 1910 to 1925. The basic idea involved in the "Atomic Energy for Military Purposes," by H. D. Smyth, Princeton University Press, 1945. separation by gaseous diffusion is the fact that gas molecules are constantly in motion and that at a given temperature the light molecules in a mixture move faster than the heavy ones. Thus, if several different kinds of gaseous molecules are contained in a box, any given light molecule will hit the sides of the box more often than any given heavy molecule. If the box is porous, more light molecules in proportion will escape than heavy ones. This is, in effect, all that is involved in the gaseous diffusion plant. Since the U-235 molecule is so very nearly the same weight as the U-238 molecule from which we desire to separate it, the amount of enrichment accomplished by each passage of gas through a porous membrane is very small. Therefore, to accomplish the desired enrichment, the diffusion of the gas through the membrane must be repeated several thousand times. Also, since there are billions of molecules in a gram of material and we are providing holes for individual molecules to pass through, we must provide billions of holes, so that the membrane area required is tremendous. Oak Ridge Plant The foregoing in part explains why the Oak Ridge gaseous diffusion plant is so large that it cost 545 million dollars to build and requires the output from a 238,000 kilowatt power-generating station. The main building of this plant is 1/2 mile long and 1/4 mile across. The several thousand diffusion stages contain many acres of diffusion membrane and consequently contain a very large volume which has to be filled with gaseous uranium hexafluoride. The amount of material in process at any one time, that is, the operating inventory, is therefore very large. The operation of the plant requires continuous information as to pressures, temperatures, pumping rates, gas analyses, and other operating variables, so that there are many thousands of instrument connections and several hundred miles of instrument tubing connected to the process piping between the many separating stages. Also, the continuous operation of such a plant with its tremendous amount of mechanical equipment requires that many and large maintenance crews be continuously at work overhauling and repairing that equipment. Thus it occurs that very frequently equipment containing uranium material is taken out of service, the material extracted, purified, and stored, later to be returned to the process. Diversion of Materials The dangers resulting from the possible diversion of concentrated product are obvious. The problem of preventing diversion of material from such a plant by use of an inspection agency which is not also the operating agency must be approached from the viewpoint that any operating agency consisting of nationals of any country might at any time act as a group to effect diversion. The term operating agency is here used to denote the organization that manages and operates the plant; it may or may not own the plant. The term inspection agency is used to denote a force of technical experts which is on duty twenty-four hours a day and which may equal the operating force in number. It is assumed that this inspection force would have the authority and responsibility to know at all times what was going on in the plant. A further point for clarification is that the present remarks are based on present-day design of the gaseous diffusion plant. Conceivably, future plants might be designed in such a way as to change greatly the problems of inspection and management. Due to the size and complexity of the present plant, the withdrawal of material by the operating agency could be done with extreme ease from any number of places by any number of methods with little risk of immediate detection by any inspection agency. Such withdrawal could be made by taking oversize samples, by storing up material in equipment which was subsequently taken out during maintenance procedures, by not immediately discovered secret withdrawal connections, or by other more or less obvious methods. Diversion of material could be concealed by an unreported improvement in the process, since recovery and operating efficiency of any process plant are continually being improved by the operating organization. Diversion could also be covered up by very minor falsifications of isotopic analyses used in calculating material balances. The detection by an inspection agency of such deliberate diversions by an operating agency would take months to accomplish, since the accuracy of a plant material balance depends on the statistical study of a large amount of data. The operator could explain such diversions by stating that any discrepancies that might show up in the material balance were due to material remaining on the inner metal surfaces of the plant. Real inspection and control can be exercised only by the operating agency and even here absolute assurance rests on the loyalty of the members of such an organization. Clandestine Plants In the case of clandestine plants of the gaseous diffusion type, the possibilities of detection would depend to some extent on the size of the installation. The construction and operation of a plant of the size of the present one at Oak Ridge would be difficult to conceal. However, it is conceivable that such a plant might be broken up into a number of smaller plants, hidden at different locations and normally |