Objectif THE MAIN OBJECTIVES HAVE BEEN TO COMPLETE THE ASSESSMENT OF THE LONG TERM CORROSION OF CARBON STEEL OVERPACKS IN GRANITE OR CLAY-LIKE FORMATIONS AND TO FINALISE THE MATHEMATICAL MODEL FOR LOCALISED CORROSION. THE VALIDITY OF THE PREVIOUSLY DEVELOPED GENERAL CORROSION MODEL AND THE IMPROVED LOCALISED CORROSION MODEL WILL BE TESTED BY THE RESULTS FROM THE EXPERIMENTAL PROGRAMME. THE OVERALL PROGRAMME WILL YIELD GENERIC MODELS FOR GENERAL AND LOCALISED CORROSION WHICH MAY BE APPLIED TO EVALUATE OVERPACK LIFE, OR THE METAL THICKNESS NEEDED TO ACHIEVE A SPECIFIC LIFE, IN GRANITE OR ARGILLACEOUS DISPOSAL SITES. The long term corrosion of carbon steel overpacks, for the disposal of nuclear waste in granitic and argillaceous environments, has been investigated. The aim of the work has been to develop mechanistic mathematical models of uniform and localised corrosion, and to test their validity by comparison with experimental data. It has been found that both models overestimate the rate of attack by factors of 4 to 6. The uniform corrosion model could be improved by including the retarding influence of accumulated corrosion products. However, in the case of localised corrosion, attention should focus on improving the estimate of the maximum period for localised attack, based on the criterion that such corrosion is only feasible under aerobic conditions. If, as seems likely, this period is only 20-40 years, the extent of penetration may then be estimated with sufficient accuracy from experimental data. However, the localised corrosion model, in its present form, will still have an important role in investigating the importance of different environmental conditions with regard to the rate of corrosion penetration.The long term corrosion of carbon steel overpacks for the disposal of nuclear waste in granitic and argillaceous environments has been investigated. The aim of the work has been to develop mechanistic mathematical models of uniform and localised corrosion, and to test their validity by comparing them with experimental data. It has been found that both models overestimate the rate of attack by factors of between 4 and 6.1. COMPLETION OF LONG TERM TESTS TO VALIDATE THE MATHEMATICAL MODEL OF GENERAL CORROSION. 2. CONTINUATION AND COMPLETION OF LONG TERM TESTS TO EVALUATE THE LOCALISED CORROSION MODEL. 3. COMPLETION OF THE MATHEMATICA MODEL FOR LOCALISED CORROSION. Champ scientifique engineering and technologyother engineering and technologiesnuclear engineeringnuclear waste managementnatural sciencesmathematicsapplied mathematicsmathematical model Programme(s) FP1-RADWASTOM 3C - Research and development programme (Euratom) on the management and storage of radioactive waste, 1985-1989 Thème(s) Data not available Appel à propositions Data not available Régime de financement CSC - Cost-sharing contracts Coordinateur United Kingdom Atomic Energy Authority Contribution de l’UE Aucune donnée Adresse 353,Harwell OX11 0RA Didcot - Oxfordshire Royaume-Uni Voir sur la carte Coût total Aucune donnée
