Corrosion of aluminum by Woncote veneer plaster

[IFRs]

Hello. I'm trapped on a job overseas and a new question came up. My internet is very spotty so I could not do an adequate search. I hate to ask but does anyone have knowledge or experience with Woncote veneer plaster with respect to corrosion on aluminum sheeting 5052-H32 welded with 4043 wire to 6061-T6. The contractor wants to mix it with water to a thin paint viscosity, paint it to some aluminum welds, allow it to dry, then apply diesel to the backside. They will wait overnight, then look for leaks in the now dry chalk. Thank you all and I apologize for asking for help before exhausting all other sources of research, etc. The Woncote may remain on the aluminum surfaces for a month or so, inside a tank on Grand Bahama Island then it should get washed off with oily liquid.

 

[btrueblood]

IFR,

Without knowing exactly the composition of the plaster, I would say to avoid contact of aluminum with wet plaster or cement. Woncote is a gypsum-based plaster, so the possibility of Calcium chloride salts as an impurity (or a reaction product forming from salt air exposure in your location) is something to worry about. The attached reference warns against wet cement/CaCl2 contact. Would it corrode enough to weaken the tank before it dried? Dunno, but I'm not much of a gambler. Why not spray or brush on a cornstarch/water or starch/alcohol mixture instead? I'm assuming you can keep your tank dry while doing the leak test.

http://www.cement.org/for-concrete-...construction/aluminum-frames-in-masonry-walls

 

[IFRs]

Let me add some clarity. I am installing an aluminum internal floating roof ( essentially a raft ) inside an oil storage tank almost 300 feet in diameter. There are about 32,000 feet of shop welded and tested seams that need a final NDE. The contractor has chosen to do a liquid penetrant test using diesel oil from the top side. They want to paint the underside at the weld with a thin slurry of woncote and let it try, leaving a strip of dust. If any liquid penetrates the weld or a crack in the base metal, it will show up more and be easier to see with the dust there. When all is finished, the tank will be filled with oil and when the floating roof begins to float, the oil will contact the underside where the dust is, washing it away. I have asked them if corn starch will work, and plan to test both at the office when I return. However, time being short they are looking for an answer quickly. I have used a spray-on indicator for dye penetrant testing, this is similar but a little home-made due to the remoteness of the project and the relative difficultiy of getting lots of specialty products.

Thanks again for your thoughts...

 

[btrueblood]

Yeah, got the idea on your first go, IFR. My worry would be not the initial application and drying, it's the idea of leaving the plaster on the surface for a long duration (month) in a high-humidity environment that may re-wet the plaster and cause corrosion over the long term. I think, if the plaster was fully removed/cleaned from the surface after the leak test, I'd be much less worried about it.

 

[IFRs]

Thanks for the input. I agree with your concerns.

Just for kicks and giggles I made a thin slurry of corn starch and applied it to aluminum samples. It dried very nicely and made a nice layer that would definitely show leaks but it did not adhere to the surface strongly enough to stay on when I manipulated the sample as it would be in the field during the diesel application.

The MSDS on Woncote says the pH is "9.3 in water" which is above the range I like to see which tops out around 8. However, given the thin layer, the short duration, that it will be submerged in oil soon, the fact that air laden moisture is better than an aqueous solution and that I can ask the field crew to wipe it off as they go, I'm inclined to run some more tests with a similar product and go forward.

Any other comments regarding either an alternative solution, possible inhibitors, how to make the starch work better, etc would be welcome!

 

[btrueblood]

Hmm...maybe the thin corrosion layer from the plaster actually improves the bonding?

Yeah, clean the plaster off as they go and figure it's good.