Introduction - Physical states transition has important issues in geomechanics. It is involved in civil engineering through concrete frost damage by in-pore water crystallization and in geophysics through the formation and dissociation of crystalline gas hydrates. - Frost durability of cementitious materials - Ice formation in cementitious materials is the cause of billions of euros in damage to concrete structures in cold and temperate regions every year as reported by the French civil engineering works supervision established in 1980 [LCPC 2003]. Two kinds of frost deteriorations exist: internal frost and scaling [Marchand 1994]. The former takes place within the whole medium, possibly resulting in micro and macro damage. The latter is a superficial damage that consists of the removal of small chips or flakes of material. It can be very harmful as it can reduce the coating of steel reinforcement. Scaling is enhanced by the presence of de-icer salts, for instance a moderate salt concentration (about 3%) is reported to result in the most damage. This explains why in temperate regions, where de-icer salts are used to keep good roads practicability, important frost defacement is observed [LCPC 2003]. The winter maintenance of civil engineering structure is being currently predicted by numerous accelerated laboratory tests simulating the climatic solicitation [AFNOR 1995; Setzer 1997]. However, these sophisticated tests are known to be expensive, long-lasting (about 3 months for a single scaling test) and to provide phenomenologically-based mechanical behaviour laws which poorly apply to the actual in-situ materials [Kaufmann 1999]. Indeed this situation lasts over several decades through lack of a holistic physically-based quantitative assessment of the stress and strain fields in porous materials submitted to freezing-thawing cycles with or without salt. - Environmental and geotechnical issues - Gas hydrates, also called clathrates, are ice-like crystalline solids. They consist of a cage-like hydrogen bonded structure of water molecules around smaller guest gas molecules: for instance methane hydrates (also called methane ice) are stable at temperatures above the normal bulk ice melting point and high pressures (from 2.5MPa at 0°C to more than 20MPa at 20°C). They occur naturally in sediments of the deep submarine continental slopes and in the subsurface of Arctic permafrost regions. Methane clathrates dissociation into a liquid and a gas mixture has recently received attention not only as a possible energy source, but also as playing a role in large continental seafloor stability (which raises a significant risk to underwater installations, pipelines, or, in extreme cases, to coastal populations through the generation of tsunamis), as well as in climate variability (methane being a powerful greenhouse gas). While gas hydrate dissociation is well-understood from thermochemical standpoint, there is little known about clathrates behaviour in oceanic sediments, in undersea continental rocks like sandstones, or more generally in porous media. - Scientific goal of this project - We want to succeed in building-up a new multi-scale physico-mechanical coupled approach that mixes wide-ranging experiments and theoretical analyses in order to better overcome cementitious materials frost durability problem. This will be the main part of this four-year project because we already have some expertise in this scientific field from both experimental and theoretical standpoints. - We will also start on understanding the physics, chemistry, and mechanics involved in the behaviour of methane clathrates saturated porous media. Based on this comprehension, the last two-year period will allow transposing the already developed experimental tools for studying methane clathrates embedded in sandstones. This will serve as a basis for better evaluating the environmental and geotechnical hazards raised by methane clathrates in oceanic sediments or

Le projet ROSA se propose d'établir l'analyse de risque du système ferroviaire dans sa globalité pour le système français et allemand à partir notamment de l'architecture fonctionnelle de l'AEIF, d'outiller cette analyse de risque à l'aide d'un logiciel permettant de visualiser les niveaux de sécurité quantifiés des fonctions et d'en évaluer les répercussions lors de la recherche d'un optimum._x000D_ Une telle analyse bénéficiant de l'expérience des réseaux majeurs en Europe permettra l'ajout de nouvelles fonctions de sécurité dans l'architecture fonctionnelle et d'en apprécier l'impact d'un objectif de sécurité donné depuis le niveau le plus élémentaire jusqu'au niveau le plus global. _x000D_ Outre à l'analyse de risque globale, le projet prévoit la présentation de l'état de l'art concernant les méthodes d'analyse coût – bénéfice visant à atteindre les meilleurs résultats de sécurité pour des niveaux d'investissements donnés._x000D_ Le projet ROSA se positionne comme la suite directe de SAMRAIL (dans lequel la SNCF, DB, INRETS et TUD ont participé), il intègre les dernières découvertes scientifiques en matière de recherche sur la sécurité ainsi que l'expérience de la SNCF et de la DB en matière de technique d'exploitation._x000D_ Ce projet fournira donc une analyse structurelle qui permettra aux industriels, exploitants, responsables d'infrastructure et autres institutions scientifiques d'introduire des modules de décomposition complémentaires. L'intention de cette approche est de faciliter la création d'outils logiciels ainsi qu'un cadre de travail scientifique qui pourrait être développé continûment, jusqu'à n'importe quelle profondeur de décomposition maîtrisée._x000D_ De plus la définition d'une analyse de risque globale du mode de transport ferroviaire pourra permettre à l'Agence Ferroviaire Européenne d'aborder la définition d'Objectifs Communs de Sécurité avec un pragmatisme garantissant le réalisme et la faisabilité des systèmes techniques jusqu'au niveau le plus bas._x000D_ _x000D_ _x000D_