Corrosion Monitoring of Steels Embedded in Swelled Bentonite Clay

Abstract

An in-situ corrosion monitoring system of metals embedded in swelled bentonite clay was developed to predict corrosion rate of nuclear waste container in the repository environment. In the monitoring system three thin iron wires (0.2 mmf in diameter, 55 mm in length exposed to bentonite) were used to measure the corrosion rate by using resistometry, impedance and DC coupling current measurements. Three iron wires and Ag/AgCl-ink electrode were embedded in a laminate sheet and attached to the bottom of the test cell filled with bentonite powder. Bentonite powder was then swelled with purified water. The cell temperature was kept at 70°C during test period of ca. one month. All operation was conducted in a globe box filled with N2gas to reduce oxygen concentration in the test environment. Dissolved oxygen (DO) concentration in swelled bentonite clay was monitored by using a luminescent DO sensor embedded in a test cell and confirmed to be lower than the detection limit of sensor (0.1 ppm). Corrosion loss of corroding iron wire was estimated by resistometry in which electric resistance change of corroding wire was measured as a relative value against the non-corroded wire for temperature compensation. Small constant AC current was applied to wire electrodes and AC voltage response at wires was measured by using a digital lock-in amplifier to detect very small change of their resistance. Impedance spectra between two corroding wires was also measured by using a FRA under non-bias condition. Coupling current between two corroding iron wires was also measured to enable evaluation of anode-cathode distribution (non-uniform corrosion) on embedded wires. In the test period bentonite powder was getting wet with penetration of water and corrosion of iron wire initiated when water reached the metal surface. This process took ca. 1000-2000 ks, depending on the density of bentonite powder. Since the water penetrating bentonite powder reached a part of iron surface, coupling condition of corroding area with the other area may be formed. Time transition of data obtained by resistometry, impedance and coupling current revealed such changes in wetting environment and corrosion progress of iron wires. Such disproportionation of corroding surface may be considerable for a large container and may accelerate the early stage corrosion in the underground repository. After reaching steady states of corrosion progress, iron wire showed corrosion rate of 19~29 μm y–1 and tended to decrease with time. The corrosion rate was considerably lower than that measured in the test condition with insufficient removal of DO from the test cell. Impedance spectra between two corroding iron wires showed gradual decrease in impedance at the initial stage of corrosion and constant low resistance at the steady state of corrosion. Charge transfer resistance of corroding iron wire in the steady state estimated from impedance spectra was ca. 1 kW. This value was considerably lower than the value predicted from the actual corrosion rate obtained for resistometry assuming that the resistance corresponded to the simple dissolution reaction. Iron surface was actually covered with corrosion film and swelled bentonite and thus the corrosion reaction is more complicated. Raman spectrum of sample surface after the corrosion test included peaks of FeS, Fe3O4, and FeOOH. This research includes portions of the results of the “Study on Performance Evaluation for Engineering Components of HLW, FY2015 and FY2016” under a grant from the Agency for Natural Resources and Energy (ANRE) in the Ministry of Economy, Trade and Industry (METI) of Japan.