The anoxic corrosion of carbon steel liners inside failed copper nuclear waste containers is dependent on the composition of the groundwater to which it is exposed. The influence of carbonate/bicarbonate, sulfate, and chloride in solutions simulating concentrated groundwaters at pH 8.9 on the composition of the corrosion products formed on A 516 Gr 70 carbon steel has been investigated at room temperature. In situ Raman spectroscopic identification of the corrosion products formed during polarization at constant potential was conducted in a spectroelectrochemical cell. Siderite was established as the main product in sodium carbonate/bicarbonate solutions, and the observance of iron carbide indicates extensive steel dissolution in this environment. In mixed carbonate, sulfate, chloride solutions, compact deposits of carbonate-containing, and to a lesser degree, sulfate-containing green rusts were formed, along with small amounts of magnetite. In chloride-dominated solutions very thin, compact films, undetectable by Raman spectroscopy, were formed. Ex situ Raman analysis suggests this film may be magnetite.One possible approach to Canadian nuclear fuel waste management involves disposal in a deep geologic repository. 1 It has been proposed that the nuclear fuel ͑UO 2 ͒ be placed in carbon steel-lined copper containers 2 and placed in tunnels within this repository. A system of buffer and backfill comprising highly compacted bentonite clay and clay mixed with crushed rock would be placed in immediate contact with the containers, thus subjecting them to hydrostatic and clay swelling pressures when groundwater eventually enters the repository. 3 Under the anoxic conditions anticipated in a Canadian repository, copper should sustain minimal corrosion damage, 1 but the potential for container failure must be considered and the consequences evaluated. Upon failure, groundwater leakage into the container will sustain corrosion of the carbon steel, resulting in the release of Fe 2+ and H 2 into the aqueous solution. A key question is whether or not the anoxic corrosion of carbon steel will lead continuously, over long periods of time, to the production of the potential scavengers, Fe 2+ and H 2 . These corrosion products would influence the redox conditions within the waste container and could suppress fuel corrosion by scavenging radiolytic oxidants.A model has been developed for fuel and carbon steel corrosion within a failed container and is published elsewhere. 4 In this model it is assumed that the copper container and any oxidizable minerals and organic material in the repository 5 has scavenged oxidants from the groundwater prior to its entering the container. As such, the sole source of oxidants to drive fuel corrosion would be the products of the ␣-radiolysis of water ͑H 2 O 2 , O 2 ͒ within the failed container. 6 The key reactions incorporated into this model are illustrated in Fig. 1. Our previous studies have shown that the chemical and physical properties of the corrosion product deposits on the steel surfac...