"304 ss Tank Bottom Failure"

Question:

"Bottom of welded 304 s/s tank containing heated phthalic anhydride (300F) failed. Not possible as yet to see the failed area. However, it appears that steam condensate mixed with phthalic anhydride (which seeped out of tank side over a period of time) may have found its way under the tank bottom which is on a concrete foundation. Could this combination possibly be corrosive to the 304 ss? After the subject tank insulation was removed it was found that there were both copper and stainless steel sheet heaters (125 psig saturated steam) located on the tank shell and they were interconnected. The heaters were strapped around the circumference with SS straps and also a thermal cement had been applied. This was originally installed 18 years ago and the cement had turned to a black powder. Most of the copper heaters were leaking and badly corroded. Could these dissimilar materials result in a galvanic action which would send stray currents through the tank wall and possibly accelerate the corrosion of the tank welds? My understanding is that liquid (molten) phthalic anhydride (say at 300F) is not corrosive to carbon steel.The failed tank was fabricated of 304ss as previously stated. Also I understand that 304 ss work-hardens when continuously flexed over periods of time resulting in an increased Brinnell hardness and greater susceptibility to ing. In the case of the failed tank the flexing would be due to the combined effects of thermal cycling,pressure cycling causing bulging when the tank is filled and a reversal as the tank is ewmptied and, perhaps, intermittent seismic loading. Since carbon steel does not work-harden, it would seem that carbon steel not 304 ss would be the material of choice for a phthalic anhydride storage tank. Could you comment on the corrosivity aspects of phthalic anhydride and 304ss/carbon steel as well as the preferred material for thi sservice?"

Answer:

Dilute, aqueous solutions of phthalic acid can be moderately corrosive (>50 mpy) to 304SS at elevated temperatures (>200F). In addition, phthalic acid can cause intergranular corrosion of 304SS which would exacerbate corrosion rates. If the tank was manufactured from regular carbon grade 304SS (0.08%), accelerated corrosion may be anticipated at welds. The shell heater galvanic couple would not have contributed to stray current corrosion of the tank floor or floor welds on either the product side or the soil side. The potential of 304SS should be slightly negative with respect to copper, therefore, if galvanic corrosion occurred it should be evident on the 304SS half of the couple next to the copper connection. Any galvanic corrosion that occurred should be on the water (I.D.) of the heater sheets unless the O.D. were continuously wet. Even then the corrosion should be confined to the immediate area of the couple based on what I am assuming was a heat transfer liquid of low conductivity. My data shows that carbon steel is rated "good" with respect to phthalic anhydride up to ~200F. Good meaning less than 20 mpy corrosion rate. Type 304SS is rated excellent at temperatures over 300F and would be a material of choice for this service. (I would specify the low carbon grade of 304SS, i.e., 304LSS).I would expect some finite corrosion of carbon steel, especially if the tank vapor space is not inert. Therefore, carbon steel would not be a good choice if color or iron contamination of the anhydride were an issue. With respect to the flexing and work hardening of 304SS, work hardening of 304SS should not appreciably increase corrosion rates. If plastic straining is occurring, it may lead to a fatigue failure, however. Also, carbon steel does work harden, just not at the same rates as 304SS.

<< return to the previous page ::