I think pitting corrosion can still occur because of the defects present in the oxide barrier layer. Halides adsorb on the oxygen vacancies and pull out cations from the inner barrier layer and it also results in cation vacancy condensation at the metal/barrier layer interface that results in pitting corrosion.

https://iopscience.iop.org/article/10.1149/1.2069096/pdf

Kind regards.

Zahed.

Dear Zahed,

thanks for your answer. I assume you are reffering to the pitting mechanism proposed by the PDM. I think this mechanism is very interesting and I am currently working on implementing it to my model of oxide film growth and breakdown.

However, as far as I know the proposed mechanism is only a hypothesis and has never been proven to be true. I have a few questions regarding the proposed mechanism:

1.) XPS studies of passivated stainless steels could show that there is no Cl- incorporated into the passive film in Cl containing solution if the surface is passivated and afterwards brought into contact with the Cl containign solution (Natishan 2010 - Chloride interactions with the Passive Film on Stainless Steel - J. Electrochem Soc.). Thus, the Schottky-Pair reaction following the Cl- injection into the film can be ruled out from the breakdown mechanism. If the Schottky-Pair reaction would be responsible for flm breakdown Cl would stay in the film. But there is also another mechanism after Cl injection, which is formation of metal-Cl complexes and their ejection to the electrolyte. (of course the XPS measurements are only integral and maybe the Cl is only incorporated to the film at weakpoints of the passive film)

2.) The number of defects to form a vacancy condensate is very high (for IronOxide there are around 150 kmol/m^3 atoms) thus the kinetics of chloride injection and metal-chloride ejection have to be very high compared to all other reeactions.

3.) I found a paper in which they performed polarization scans on defect free stainless steel surfaces and there was absolutely no pitting in the potential range between OCP and O2 evolution (Nishimoto 2020 Passivity of (Mn,Cr)S inclusions Corrosion Science). If the PDM mechanism is ture - I totally agree with you at some point every oxide film containing oxygen vacancies will break due to Cl injection followed by cation ejection.

What do you think?

Dear Dr. Bösing,

As far as I know based on the PDM mechanism for the pitting corrosion in the presence of chloride, chloride does not inject into oxide or incorporate into the body of the oxide inner layer but it physically absorbs on the oxygen vacancies and eject metal cation form the oxide surface. So, XPS will not detect any chloride in the oxide layer. Figure 10 of the paper I sent the link in the previous message shows that vacancy condensation occurs between metal/film interface. Also, pitting of metal in the presence of chloride depends on the concentration of chloride as well.

https://iopscience.iop.org/article/10.1149/1945-7111/ac580c/meta

Article Studies on Pitting Corrosion of Al-Cu-Li Alloys Part II: Bre...

Kind regards,

Zahed.

I think as soon as chloride adsorbs at the oxygen vacancies at the film surface it can either eject metal from the film by formation of a complex or it can diffuse into the film. Since the charge of a chloride ion "sitting" at a oxygen vacancies is +1 the electric filed will push the Cl ions towards the film/solution boundary the flux wont be huge but I think it is not impossible and dependign on the Cl concentration and the ejection kinetics.

Dear Dr. Bösing,

To the best of my knowledge, chloride ions (with a size of 167 pm) are too large to diffuse readily through a densely packed inner oxide layer. However, they can diffuse through the more porous outer layer. It's worth noting that even oxygen anions, which are interstitial ions, and have smaller size compared with chloride ions have not been taken into account in the PDM due to their relatively large size (140 pm).

I agree that a diffusion through the crystal would be unlikely. But how about a diffusion throug the crystal defects (as diffusion in crystal normaly occur). Thus, the Cl ions are transported slowly from vacancy to vacancy?(However, I agree this transport would be rather slow and might be neglectable).

To my knowledge in other oxides (Al and Ni for example) Cl could be found in the compact barrier layer

Dear Dr. Bösing,

I think the only possible way for chloride to diffuse through an inner layer oxide is as an interstitial which is highly unlikely occurs due to its large size. I am not aware of the diffusion of chlorides through other oxides such as Al and Ni.