Atmospheric Corrosion of Copper in Contact with Bentonite with and without Deliquescent Salts

Canada's high-level radioactive waste disposal plan entails using tightly sealed used fuel containers (UFC) emplaced in a deep geological repository (DGR). The UFC consists of a carbon steel vessel for mechanical support and an outer 3 mm copper coating for corrosion protection. After emplacement in a DGR, containers will contact highly compacted bentonite clay. The DGR conditions will be initially warm and humid (up to 90°C), due to the radioactive decay processes within the used fuel, and oxygen will be present because of air trapped during the placement process. During this period, the heat produced by radioactive decay will cause sections of the bentonite to become desiccated, leading to shrinkage. This will create gaps between the bentonite and the copper surface, with some bentonite residue remaining attached to the copper surface. Once this surface becomes exposed to humid air, non-uniform atmospheric corrosion will be possible since bentonite contains traces of mineral salts. This study examined the effect of different forms of bentonite—highly compacted bentonite (HCB), bentonite residue (BR, particles left after HCB contact), and bentonite slurry (BS)—on copper corrosion. To amplify the impact of different forms of bentonite, a layer of sodium chloride (NaCl) was also applied to the copper surfaces before adding bentonite. Thus, this series of experiments investigated copper corrosion under two conditions: (part 1) in direct contact with various forms of bentonite; and (part 2), in contact with a layer of NaCl beneath the coverage by the different forms of bentonite. In Part 1, atmospheric corrosion of O-free, P-doped wrought copper (SKB-Cu) and cold-sprayed copper (CS-Cu) were studied with different forms of bentonite over various exposure periods. In Part 2, CS-Cu samples were studied with a NaCl layer beneath the bentonite deposit over similar periods. All samples were exposed to humid air at 75°C and 75% relative humidity (RH). Based on results obtained from optical microscopy and scanning electron microscopy (SEM), Cu samples covered with BR, HCB, and BS exhibited localized, uniform, and non-homogeneous corrosion, respectively in part 1. The corrosion features of both SKB-Cu and CS-Cu were very similar. After applying NaCl (part 2), the morphology of corrosion products showed significant changes. For both SKB-Cu and CS-Cu in part 1, Raman spectroscopy and X-ray photoelectron spectroscopy (XPS) showed that the corrosion products formed on the surfaces covered by either HCB or BS were mainly CuO, Cu(OH) 2 , and Cu 2 O, while on the BR-covered samples, only Cu(OH) 2 and Cu 2 O formed. For part 2, the corrosion products on all samples consisted of Cu 2 O, CuO, and Cu(OH)₂, though their distribution was uneven after applying NaCl. After a 5-month corrosion experiment (part 1), focused ion beam-scanning electron microscopy (FIB-SEM) images showed a significantly thinner corrosion layer (40 nm) on the samples exposed to HCB than on those exposed to other forms of bentonite (400 nm). Samples covered with HCB exhibited a lower corrosion rate (around 0.06 µm/yr) than samples exposed to the other forms of bentonite (around 0.6 µm/yr). After removing corrosion products by pickling, most samples displayed uniform shallow pits. In part 2, based on weight loss measurements on specimens exposed to humid air at 75°C and 75% RH for 5 months, BS-covered samples showed a lower corrosion rate (around 1.2 µm/yr) than those exposed to other bentonite forms (6–9 µm/yr). After corrosion product removal, most samples displayed a rough, uneven surface, with BS-covered samples showing a smooth surface and uniform shallow pits. In Part 1, no significant difference was found between SKB-Cu and CS-Cu corrosion behavior, with HCB-covered samples showing the lowest corrosion rate. In Part 2, after salt printing, the overall corrosion was more severe, with BS-covered samples exhibiting the lowest corrosion rate. In both parts, the corrosion rates of all samples decreased over time.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.