Numerical Predictions of Dry Oxidation of Iron and Low-Carbon Steel at Moderately Elevated Temperatures

Abstract

Wrought and cast low-carbon steel are candidate materials for the thick (e.g. 10 cm) outer barrier of nuclear waste packages being considered for use in the potential geological repository at Yucca Mountain. Dry oxidation is one potentiaI degradation mode for these materials at the moderately elevated temperatures expected at the container surface, e.g. 323-533 K (50-260 'C). Therefore, numerical predictions of dry oxidation damage have been made based on experimental data for iron and low-carbon steel and the theo~of parabolic oxidation. A numerical approach employing the Forward Euler method has been implemented to integrate the parabolic rate law for arbitrary, complex temperature histories. Assuming growth of a defectfree, adherent oxide, the surface penetration of a low-carbon steel barrier following 5000 years of exposure to a severe, but repository-relevan~temperature history is predicted to be only about 0.127 mm, less than O.13% of the expected container thickness of 10 cm. Allowing the oxide to span upon reaching a critical thickness increases the predicted metal penetration values, but degradation is still computed to be negligible. Based on these physically-based model calculations, dry oxi&tion is not expected to significantly degrade the performance of thick, corrosion allowance barriers constructed of low-carbon steal.

Effect of chloride content of molten nitrate salt on corrosion of A516 carbon steel.

“…Literature values of the activation energy for oxidation in air are about 25 kcal, though these values were derived from relatively short-term experiments during which oxidation followed parabolic kinetics. [16][17][18] Henshall notes that this value of the activation energy is consistent with grain-boundary diffusion of Fe ions in Fe 3 O 4 .…”

Identification and Assessment of Material Models for Age-Related Degradation of Structures and Passive Components in Nuclear Power Plants

“…For the temperature of interest for this method to be applicable, Fe 3 O 4 (magnetite) is (reasonably) assumed to be the only corrosion product, leaving out the thermodynamically unstable FeO (wüstite) and the less substantial Fe 2 O 3 (hematite). In addition, the following assumptions were also made by Henshall [1996] when developing this model:…”

“…In the literature, ݇ p varies considerably and a conservative value 4.3× 10 -5 cm 2 /s was used by Henshall [1996] as this value is the largest found in the literature. It was also reported that Q appeared to be constant at about 104 kJ/mol in the parabolic regime and at relatively low temperatures in a range of 523 to 736 K. Lower values of Q have been reported for higher temperatures.…”

Engineered materials characterization report, volume 3 - corrosion data and modeling update for viability assessments

“…The model favored for dry oxidation of the CAM (A516 Gr 55) was developed by Henshall (Henshall, 1996b) where Ap/At is the apparent penetration rate (um/yr), T is the temperature PC), and CNaais the equivalent concentration of NaCl (wt%) Note that the "equivalent NaCl concentration" is used as an independent variable in correlations This parameter is defined in terms of the concentration of free chloride, as follows (Eq 2 8-2b)…”

“…The maximum observed rate was about 300 pm/yr. In(dp/df) In(dp/df) In(dp/dij In(dp/dQ lOOO/-r Engineers are exploring several mechanisms to delay corrosive attack of the CAM by dripping water, including drip shields and ceramic coatings Ceramic coatings deposited with HVOF have exhibited a porosity of only 2% at a thickness of 0 15 cm A model has been developed to account for the inhibition of corrosion by these coatings It is assumed that the overall mass-transfer resistance governing the corrosion rate is due to the combined resistances of ceramic coating and interfacial corrosion products Two porosity models (simple cylinder and cylinder-sphere chain) are considered in estimation of the mass-transfer resistance of the ceramic coating It is evident that a substantial impedance to 0, transport is encountered if pores are filled with liquid water It may be possible to use sealants to eliminate porosity Spallation (rupture) of the ceramic coating is assumed to occur if the strain introduced by corrosion products at the ceramic-CAM interface exceeds fracture strain of the coating During the initial period of dry oxidation, any porosity in the ceramic coating is assumed to be filled with gas If the porosity is interconnected, the impedance to Oz transport and oxidation is believed to be relatively insignificant In such a case, a good approximation is to simply apply the dry oxidation rate provided by Henshall (Henshall, 1996b) The impact of dry oxidation underneath the porous ceramic is believed to be relatively insignficant based on preliminary calculations During the period of humid-air corrosion @MC), pores may be filled with either gas or liquid If porosity is interconnected and gas-filled, the ceramic coating may not significantly impede the rate of HAC. However, the Kelvin effect can lead to condensation in very small pores For example, at 100°C and 99% RH, 0 07-nm pores would be expected to be liquid filled.…”