Materials selection should strike an appropriate balance among capital, operating and maintenance costs. It should consider control of process conditions (temperature, pressure and velocity) and stream composition, use of protective coatings and additives, as well as preventative maintenance and inspection programs. The exact mix depends upon the constraints of the process. It’s a given, though, that some corrosion is inevitable at nearly every plant. After all, few materials are completely invulnerable to the conditions they’ll face in a process.
Here, we’ll look at a new plant to illustrate how numerous elements may affect materials selection. This plant is for a first-of-a-kind process that involves using a couple of organic acids to produce a high-value intermediate. While the corrosion behavior of one of the organic acids is well understood, that’s far from the case for the second. In addition, one of the streams contains a rarely encountered but potentially corrosive chemical that essentially has unknown behavior.
Initial materials selection focused on a well-known proprietary alloy — but getting it in all the forms needed for the equipment is extremely difficult. This problem has re-opened the question of what are the right materials to choose.
First, you must determine which materials are suitable. This involves understanding the corrosion rate, the failure type and the consequences of failure — and then picking a material that gives the plant a reasonable life. In one case, a client specified a plant life of only eight years but, more typically, the stipulated life is 20–25 years. The right choice for eight years may differ markedly from the one for 25 years. Failures from corrosion can result from gradual deterioration of an entire surface; steady but localized pitting; or sudden attack. Process temperature or pressure can induce failures. You also must consider the consequences of failure. Leaking a little cooling water may pose a minor issue but a sudden loss of containment of a flammable or toxic chemical may lead to catastrophic consequences. For the new plant, many alloys would be suitable, with an alloy’s molybdenum content likely a key issue for the organic acids.
Second, you must evaluate the cost balance between the available choices. In this case, a material that’s almost sure to work cost over eight times more than the least expensive one that might work. The economics of the project could tolerate a material only ≈2.5–3 times costlier than the least expensive option. This restricted the choices to six alloys.
Third, you must take into account material availability — looking at two factors: the forms in which a material is manufactured; and whether these are readily purchasable or require a special order. So, you must assess your process requirements. Do you need tubes, plate, forged fittings, thin or thick sheets? Also, how will the equipment be assembled? In this case, we could get the forms necessary but only three of the alloys had off-the-shelf availability; these were the less expensive (good) but probably lower-performing (bad) choices. We could purchase equipment in these three without excessive lead-times. In addition, these choices would simplify future spare-parts procurements and maintenance. The other three materials had regional variations in availability; one was easier to get in the United States and the other two more readily available in Europe. While this isn’t a major problem, it is a factor to consider.
Because this is a new process, the three candidate materials — the cheapest that might work; an off-the-shelf material that likely will offer much higher corrosion resistance; and an uncommon (and expensive) material expected to give excellent corrosion resistance — will undergo testing. The program will use mixtures of the organic acids to check corrosion rates and look for unexpected problems. The results should provide a prudent basis for economic and safe materials selection. (For some caveats about the use of stainless steels, see “Don’t Put Peddle to the Metal.”)
Ultimately, we’ll probably opt for a mix of materials, with the choice depending upon the equipment type and the part of the process. For example, structured packing in distillation towers is thin and corrodes on both sides; a loss of just 1–2 mils may destroy the packing. In contrast, adding a corrosion allowance to a vessel wall usually is straightforward.