99 Diseases of Pressure Equipment: Sulfuric Acid

Unlike NAC where we’re still on a learning curve, the knowledge of corrosion by sulfuric acid has not changed much in the last quarter century, and there are many good references for it included in API RP 571(1). In the refining industry the most common place to experience sulfuric acid corrosion is in the sulfuric acid alkylation units where process control or water contents of acids deviates from design operation. Some operators of sulfuric acid alky units have long believed that because of the sulfuric acid catalyst, that leaks are just part and parcel with the operation of such units. We now know that that just isn’t true. With rigid control of acid concentration, fluid velocity, temperature and the avoidance of acid carryover, these units can be operated reliably with high integrity.

While much of a sulfuric acid alky plant can be constructed cost effectively out of carbon steel, there are a few places where upgrades to 316L, Alloy 20, and even Alloy B-2 and C276 may be needed. As with HFA alkylation units, sulfuric acid carryover into equipment and piping systems that were not constructed to resist sulfuric acid at certain concentrations and temperatures can lead to higher than design corrosion rates. Reactor effluent lines, reboilers, deisobutanizer overheads, caustic treating systems, and flare lines seem to be most vulnerable.

Though corrosion from sulfuric acid is often general thinning, it can also be highly localized to weldments (especially HAZ), crevices and liquid vapor interfaces, as is often the case in flare lines because of the ingress of oxygen. It is not uncommon to see a knife-like line of corrosion at the tops of lines carrying concentrated sulfuric acid and at elbows, especially where maximum allowable velocities are exceeded. These types of localized corrosion are difficult to detect with point ultrasonic thickness measurements and require volumetric examinations such as raster pattern UT “B” or “C” scans.

When you do RBI on sulfuric acid containing systems, does the RBI team discuss all the possibilities for where corrosion might accelerate due to temperature excursions, fluid velocity increases, and acid concentrations below design minimums?