Heavy water and nonproliferation : topical report
The following report is a study of various aspects of the relationship between heavy water and the development of the civilian and military uses of atomic energy. It begins with a historical sketch which traces the heavy water storyfrom its discovery by Harold Urey in 1932 through its coming of age from scientific curiosity to strategic nuclear material at the eve of World War II and finally into the post-war period, where the military and civilian strands have some- times seemed inextricably entangled. The report next assesses the nonproliferation implications of the use of heavy water- moderated power reactors; several different reactor types are discussed, but the focus in on the natural uranium, on- power fueled, pressure tube reactor developed in Canada, the CANDU. The need for and development of on-power fueling safe- guards is discussed in some detail. Also considered is the use of heavy water in plutonium production reactors as well as the broader issue of the relative nuclear leverage that suppliers can bring to bear on countries with natural ura- nium-fueled reactors as compared to those using enriched designs. The final chapter reviews heavy water production methods and analyzes the difficulties involved in implemen- ting these on both a large and a small scale. It concludes with an overview of proprietary and nonproliferation constraints on heavy water technology transfer. Our major conclusions are as follows: 1. On-power fueling of CANDU reactors leads to special, well recognized safeguarding problems. These have been addres- sed by a safeguards development program, encompassing both systems analysis and hardware development, jointly sponsored by Atomic Energy of Canada, Ltd. (AECL), the Canadian Atomic Energy Control Board (AECB), and the International Atomic Energy Agency (IAEA). The approach involves surveillance, containment, and item-counting of irradiated fuel bundles. Although the complete system has not as yet been tested on an operating reactor, it appears to be a good example of "proliferation resistance engineering." The major problem may be the political one of obtaining agreement to incorporate the system in operating reactors and those under construction. 2. The question of relative leverage on natural uranium vs. enriched uranium fuel cycles does not have a neat answer. At the moment, most of the countries of proliferation concern have neither large amounts of uranium ore nor the ability to enrich it. (There are, of course, some significant excep- tions, the most obvious being South Africa which has both.) In the near term, the chances of achieving a consensus among current suppliers of separative work, all of whom belong to the London Club, not to supply it in the event of violations of nonproliferation agreements, also seems greater than the prospects of reaching a similar agreement among all countries who might be able to supply uranium ore. If we assume in ad- dition that the malefactor also can produce heavy water--no small matter--the potential leverage advantage would seem to lie with enriched reactors. On the other hand, the spread of enrichment technology--which is easier to rationalize on civil- ian grounds if enriched reactors are in place--could tip the scales the other way. In general, however, this weighing of enriched vs. natural uranium fuel cycles is unnecessarily restrictive. Experience has shown that there are many poten- tial levers--nuclear and non-nuclear--which can be used to persuade countries to adhere to nonproliferation norms. The heart of the matter is the political will to use these in the face of conflicting policy objectives. 3. Unlike uranium enrichment via gaseous diffusion and the gas centrifuge, key aspects of which are closely held on nonproliferation grounds, techniques for heavy water produc- tion, particularly by hydrogen sulfide-water exchange (the GS process), have been extensively documented in the open liter- ature. Nevertheless, construction and operation of large plants are difficult, and thus there is good reason to believe that the technology will not spread rapidly through the indig- enous efforts of developing countries. Unlike uranium enrich- ment and fuel reprocessing, heavy water production does not provide a direct route from civilian fuel cycle to weapons- usable materials; on these grounds a logical quid pro quo forits transfer would be adherence to the Non-Proliferation Treaty (NPT) or acceptance of the principle of full-scope safeguards by the recipient.
MIT Energy Laboratory
Heavy water reactors, Deuterium oxide, Nuclear nonproliferation
The following license files are associated with this item: