Oxide Film Behaviour on A508III Steel and 309L/308L Cladding in High-Temperature Water: A Study Note
Overview and Background
This paper from the Shanghai University corrosion group addresses a question that sits at the heart of primary circuit integrity in pressurised water reactors: what happens to the oxide film on the low-alloy steel pressure vessel and its stainless steel cladding when the water chemistry changes? The authors exposed A508III steel and 309L/308L weld overlay to high-temperature water under two conditions—oxygenated and deoxygenated—and characterised the resulting surface films. For anyone who has signed off on a reactor pressure vessel, this is not an academic exercise. The oxide film is the first line of defence against corrosion, and its stability determines whether the vessel survives its design life or fails prematurely. The study's focus on the difference between 309L and 308L is particularly valuable, because it challenges the assumption that all austenitic cladding behaves the same way.
Key Findings and Technical Interpretation
The results are clear and, in my experience, consistent with what we see in the field. In oxygenated high-temperature water, the A508III steel surface forms γ-Fe₂O₃, a haematite-type oxide that is relatively loose and non-protective. In deoxygenated water, the surface forms Fe₃O₄, a magnetite spinel that is dense, adherent, and far more protective. This is a fundamental difference: the oxygen content in the water directly dictates whether the base metal corrodes slowly or rapidly. For the cladding, the trend is reversed in a subtle way. In deoxygenated water, the 309L/308L overlay develops spinel-type oxide particles that are larger and denser than in oxygenated water. Lower oxygen content promotes the formation of this protective spinel, which is exactly what we want from a corrosion barrier. The paper also reports that 309L surfaces show pitting in both conditions, while 308L surfaces remain pit-free. This is the most striking observation, and it deserves careful attention.
Metallurgical and Practical Implications
Why does 309L pit when 308L does not? The authors attribute it to the preferential dissolution of inclusions in 309L. In my experience, this is a classic inclusion-driven pitting mechanism. 309L has a higher chromium content (typically 23–25% Cr) than 308L (18–20% Cr), which should improve corrosion resistance. However, the weld overlay process—whether it is strip cladding, SAW, or GTAW—inevitably introduces non-metallic inclusions, particularly manganese silicates and chromium-rich oxides. These inclusions act as preferential dissolution sites in high-temperature water, and once they dissolve, they leave behind a pit that can grow. The fact that 309L shows this behaviour while 308L does not suggests that the inclusion population in 309L is more susceptible, possibly due to the higher Cr content promoting different oxide inclusions or a different solidification structure. From a practical standpoint, this means that the choice of cladding material is not simply a matter of alloy composition; it is also a matter of inclusion control. If I were specifying a cladding for a PWR vessel, I would insist on strict inclusion limits and a low-oxygen steelmaking practice, regardless of whether 309L or 308L is used.
| Condition | A508III Base Metal Oxide | 309L/308L Cladding Oxide | Pitting Behaviour |
|---|---|---|---|
| Oxygenated water | γ-Fe₂O₃ (loose, non-protective) | Spinel-type (smaller, less dense) | 309L pits; 308L does not |
| Deoxygenated water | Fe₃O₄ (dense, protective) | Spinel-type (larger, denser) | 309L pits; 308L does not |
Engineering Practice and Lessons Learned
This study reinforces several lessons I have learned the hard way over the years. First, water chemistry control is not optional. In a PWR, the dissolved oxygen content must be kept low—typically below 5 ppb—to maintain the protective magnetite film on the base metal. If oxygen levels spike, the base metal will begin to form haematite, and the corrosion rate will increase by orders of magnitude. Second, the cladding material selection must be based on more than just the alloy composition. The inclusion population, the weld process, and the post-weld heat treatment all play a role in determining whether the cladding will pit. Third, the difference between 309L and 308L is a reminder that we cannot assume all austenitic stainless steels behave identically. Each alloy has its own microstructural fingerprint, and that fingerprint determines its corrosion performance. Finally, the paper highlights the importance of understanding the oxide film as a dynamic system. The film that forms in one condition may not be the film that forms in another, and the transition can be abrupt. For plant operators, this means that any change in water chemistry must be carefully managed, because the oxide film will respond quickly and not always favourably.
Summary and Reflections
This is a well-executed study that provides clear, actionable insights for anyone involved in the design, fabrication, or operation of PWR primary circuit components. The key takeaway is that oxygen content in high-temperature water is the dominant factor controlling oxide film composition and morphology on both the base metal and the cladding. The finding that 309L is susceptible to pitting while 308L is not, under identical conditions, is a critical piece of information that should influence material selection decisions. In my own practice, I would use this data to justify a preference for 308L cladding in critical applications, or at minimum to impose stringent inclusion controls on 309L. The study also serves as a reminder that corrosion is a system property, not just a material property. The water chemistry, the temperature, the stress state, and the microstructure all interact, and we must understand all of them to ensure long-term integrity. For young engineers, this paper is an excellent example of how a focused laboratory study can illuminate a real-world problem and provide guidance that is directly applicable to the shop floor and the plant.
Zhuojin Pipe Fitting Co., Ltd