The first European industrial-scale tests of MOX took place in 1963 at the BR3 reactor in Mol, Belgium and in 1974 at the Chooz A reactor (now shut down) located on the French border with Belgium. These tests were the result of early French-Belgian MOX development. Belgonucléaire and COGEMA began producing MOX fuel jointly at two small plants in Dessel, Belgium (which started operation in 1973), and Cadarache, France (which started operation in 1970). The capacities of these plants are 35 metric tons and 15 metric tons of MOX fuel per year respectively. Also as a part of the joint French-Belgian work, four Belgian pressurized water reactors (PWRs) at Tihange and Do'l and the 28 first French 900 MW PWRs, which began operation between 1980 and 1984, were designed to use MOX fuel. They have four unused vessel head guide tubes in which supplementary control rods deemed necessary for MOX use can be placed. Curiously, the twenty-two 1300 MW reactors that were built later in France are not adaptable to MOX use. This is probably because after an initial wave of MOX development for LWRs, emphasis for plutonium fuels shifted fast neutron reactors (or "breeder" reactors), and MOX use in light water reactors was relegated to second place.

It was not until 1984, after it became clear that initial hopes for breeder reactor programs would not be achieved, that Belgonucléaire and COGEMA were able to regroup to commercialize MOX in LWRs. As a result of these new efforts, MOX was loaded into a French reactor for the first time in 1987, at Saint-Laurent-les-Eaux (in the Loire region). Out of the 52 fuel assemblies replaced each year (one-third of the entire core), 16 are MOX. Since then, nine other reactors have also been loaded with up to 30% MOX, the maximum loading accepted by French nuclear safety officials for MOX, containing no more than 5.3% plutonium. In Belgium, as a result of parliamentary debates in December 1993, two reactors can be loaded with only up to 20% MOX core, but with a 7.7% plutonium content. This second wave of MOX development for LWRs also resulted in a new fabrication facility in Marcoule, France, called Melox, with a capacity of 115 metric tons. A license was granted for construction of the plant in 1990, and it began operation in 1995. By the end of 1996, it had reportedly delivered 96 MOX assemblies to the French utility, EDF.

Only 16 of the 28 French 900 MW reactors were licensed to receive MOX at the time of their construction. Public debates are currently being conducted which would allow MOX use in four additional reactors in Chinon, on the Loire river. This step runs counter to the government decision to have experts conduct an "environmental and economic assessment of MOX use" by June 1997. The Forum Plutonium has also demanded that results of the public survey in Chinon be reported in autumn 1997.

The security, safety, and economics of MOX fuel use have long been questioned by many experts in France. In November 1990, when the decision was made to build the Melox plant, Jean-Paul Schapira, a well-known nuclear physicist questioned the value of MOX use in the journal La Recherche.¹ Recently, in the Gazette Nucléaire, Monique Sené of the GSIEN (Association of Scientists for Nuclear Energy Information), shows that the objections raised by J.P. Schapira have been borne out by the MOX fuel assemblies that have been used by EDF.²

Schapira and Sené identify a number of safety problems posed by MOX compared to traditional uranium fuel which it replaces: more delicate fabrication of fuel rods to protect against contamination, greater risk of loss of control during reactor operation despite the presence of extra control rods, release of fission gases, corrosion of fuel rods during reactor operation. Given the signs of aging which are now appearing in the French 900 MW reactors, these complications are all the more problematic.

The security problems associated with MOX use are linked to the transport of nuclear materials that can be used to make nuclear weapons or other weapons that could disperse radioactive material. In France, plutonium and MOX fuel are transported by road under police escort, during the day only, along routes which are kept secret.³ Since plutonium is produced at La Hague (in the northwest Contentin region), and the MOX fabrication facilities are in Belgium and in southeast France, MOX fabrication requires a significant number of plutonium shipments. In addition, the transport of MOX along routes scattered throughout France and Europe, creates the potential for a radiological pollution of the ecosystem that could last for millenia.⁴ In addition, MOX is intimately related to the policy of reprocessing spent fuel, which is probably the most environmentally dangerous activity of the nuclear industry. Recent studies near La Hague and Sellafield have shown numerous health and environmental problems resulting from reprocessing.

At the same time that plutonium is accumulating at La Hague (36 metric tons at the end of 1995), the price of natural uranium is decreasing, and uranium produced from reprocessing is also accumulating. By June 1, 1995, 7500 metric tons of uranium had been recovered at La Hague from reprocessing of spent fuel, enough to fabricate 15,000 fuel assemblies. Given these large amounts of uranium available at low prices, MOX fuel cannot compete economically with uranium fuel. MOX fuel fabrication is considerably more expensive than uranium fuel, even if the plutonium is considered to be free. If reprocessing costs are also taken into account, it becomes clear that MOX fuel is not economically viable.

MOX producers are faced also by a number of technical constraints for the fabrication and storage of MOX, which cannot but increase their cost.

• the presence of strong alpha-emitters and of americium-241, a highly radioactive gamma-emitter;

• limited storage period of 2 to 3 years for plutonium extracted for the production of MOX before its use;

• a greater enrichment of fuel necessary in order to increase the time fuel rods can remain in a reactor, which is avidly sought by EDF: 4.2 percent for uranium fuel, but 8 percent for MOX. The MOX currently authorized in France contains only 5.3 percent plutonium. It produces 30,000 megawett days per metric ton of heavy metal while uranium fuel produces 47,000. Therefore, EDF has requested authorization (so far unsuccessfully) to increase the plutonium content in MOX to 7 percent.
• tests of reprocessing MOX fuel have produced a form of plutonium that is less fissile (and therefore produces less energy) with a higher level of transuranic elements (and thus a higher level of radioactivity) than reprocessed uranium fuel. In August 1996, EDF announced that it wanted to store spent MOX fuel. Thus, so far there is no policy regarding what to do with MOX spent fuel.

  In sum, MOX, regarded by some as a way to reduce plutonium stocks, is not without its problems. Since the failure of "breeder" reactors, it remains the last chance for plutonium proponents in their competition with uranium proponents within COGEMA. However, if the mediocre economic balance-sheet for MOX were added to the plutonium industry's already disastrous environmental balance-sheet, plutonium could become a waste in France. There is, therefore, some hope that future generations will have less plutonium to manage than what is currently envisioned in COGEMA's reprocessing contracts.

  Jean-Pierre Morichaud, a retired physical-chemistry engineer, began his career at the Saclay research center in 1957. He was the president of a coalition opposed to the Melox plant from 1992 to 1994. He is currently the coordinator of the Forum Plutonium, a coalition of organizations in France, Belgium, and Switzerland.

1. J. P. Schapira, "Une nouvelle stratégie pour le plutonium," La recherche, No. 226, November 1990.

2. M. Sené, "Dossier MOX," La Gazette Nucléaire, No. 155/156, January 1997.

3. M. Pavageau, J. Hazeman, M. Schneider, Les transports de l'industrie du plutonium en France, WISE-Paris, 1995.

4. Plutonium: Deadly Gold of the Nuclear Age, IPPNW/IEER, 1992, French edition in Médecine et Guerre Nucléaire, Vol. 8, No. 3, 1993.