Know the Limits of Duplex Stainless Steels

A distillation column shows that stress corrosion cracking can occur

By Helga Leonhard and Gernot Grötsch, TÜV SÜD Chemie Service

1 of 2 < 1 | 2 View on one page

In the chemical industry, pressure vessels such as distillation columns must satisfy special requirements regarding safety and availability. Given this, they are made from high-strength materials such as duplex stainless steels. Such steels include both austenitic and ferritic phases and, therefore, have long been considered resistant to stress corrosion cracking.

However, under certain conditions — mostly involving chloride-containing fluids and elevated temperatures — even duplex stainless steels are vulnerable to this type of corrosion. This makes thorough failure analysis and corrosion testing imperative. The current condition of the materials used is not only important for safety and availability but also crucial for the profitability and the service life of existing plants.

At a chemical plant, a failure occurred in a distillation column that had been in continuous operation for ten years. The column was made from robust austenitic steel (1.4571, titanium stabilized). Stress corrosion cracking in various areas of the wall caused the failure. The bottom of the column suffered the most severe cracking, but stress corrosion also was found in upper parts of the unit at a height of up to 15 meters. Corrosion was attributed to an acidic oxygen-free aqueous chloride solution with highly oxidizing admixtures that had a temperature of 125°C in the bottom and around 90°C in the upper part of the column.

The experts sought an alternative material that is completely immune to stress corrosion cracking. After laboratory analyses, they selected a duplex stainless steel (1.4462) and reconstructed the column in this material (Figure 1). To improve the column’s resistance to stress corrosion cracking even further, its internal surface was treated by applying strong pickling agents to remove a specific amount of material (normally 3–5 µm) from the surface together with all superficial residual stresses.

The column was placed into service and examined for signs of corrosive attacks during each maintenance shutdown. Five years later, the bottom of the column — which had been in direct contact with the fluid — showed the first signs of corrosive attacks. The inspectors found pronounced pits on the inner surface of the metal wall and corrosion grooves along the weld seam (Figures 2 and 3). Electrochemical tests performed using the chloride solution from the bottom part of the distillation column as test fluid revealed the probability of pitting corrosion to be high at temperatures as low as 90°C. Given this, the experts assumed pitting corrosion to be the problem in the case at hand.

Over the following years, they monitored and documented the progress of corrosion. They marked selected surface areas and produced surface replicas on film to see whether the area affected by corrosion was growing. In addition, they measured the depth of the most-pronounced defects. A comparison of pit depths with the surface replicas produced at various times revealed that no new pits had developed and existing pits had not become either deeper or bigger. The case at hand obviously was caused by one-time corrosion that did not progress. Pitting corrosion can be tolerated up to a certain degree. However, in combination with tensile stress, it may cause stress corrosion cracking that could significantly reduce component integrity.

After another four years of operation, the distillation column had to undergo a gas pressure test. For this reason, to be on the safe side, the experts decided to re-examine some of the corrosion grooves at the circular bottom weld and the longitudinal weld of the bottom part. Dye penetration testing did not produce a clear result. Therefore, the experts carefully sanded and polished the weld surface. To exclude cracking, they repeated the dye penetration test, which now revealed cracks that had been invisible previously (Figure 4). Further measurements of crack depth showed that some cracks had penetrated the entire wall.

1 of 2 < 1 | 2 View on one page

Join the discussion

We welcome your thoughtful comments. Please comply with our Community rules.
All comments will display your user name.

Want to participate in the discussion?

Register for free

Log in for complete access.


No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments