Under-deposit corrosion in a 316L cooling-water exchanger
A 316L stainless steel shell-and-tube cooler has leaked after 14 months at a tube-to-tubesheet roll joint. The cooling water contains 900 mg/L chloride, enters at 30 C and leaves at 55 C; the tube wall on the hot side reaches 85 C. Water velocity in the tubes is 0.4 m/s, and the tubes carry a brownish deposit near the hot end. The water is treated with a phosphate-based scale inhibitor but has no filtration. The tube-side design pressure is 6 bar, and the shell side contains a non-hazardous process fluid. Inspection shows branched transgranular cracks initiating beneath the deposit at the roll joint, with adjacent pitting. The plant wants to return the exchanger to service quickly and is considering replacing the 316L tubes with 2205 duplex or titanium Gr 2, or simply increasing inhibitor dose.
1. Identify the most likely failure mechanism and explain why the combination of chloride, temperature, deposit and low velocity matters.
Read the worked answer
The most likely mechanism is chloride pitting that has initiated under the deposit and progressed into chloride stress-corrosion cracking at the roll joint. The branched transgranular morphology is characteristic of chloride SCC in austenitic stainless steel, and the pits underneath the deposit provide initiation sites. The roll joint adds residual tensile stress, which is the mechanical driver for SCC.
The combination is aggressive: 900 mg/L chloride is well above the level normally considered safe for 316L in stagnant or deposit-covered areas; 85 C at the hot wall is high enough to accelerate both pitting and SCC; low velocity of 0.4 m/s allows deposits and crevices to form; and the phosphate inhibitor does not remove the deposit or eliminate chloride. The critical pitting temperature of 316L is effectively exceeded in the deposit-covered hot zone. Simply increasing inhibitor dose would not fix the geometry or the chloride level and could even mask the problem while cracking continues beneath deposits.
2. Evaluate the material alternatives and recommend a replacement and design changes.
Read the worked answer
316L should not be reused for this duty without a major change in water chemistry or tube-wall temperature. Duplex 2205 has a much higher pitting resistance equivalent number and better chloride SCC resistance than 316L, but at 900 mg/L chloride and an 85 C wall it is still not a guaranteed choice, especially under deposits. Superduplex 2507 or a 6% Mo austenitic alloy would give more margin, but titanium Gr 2 is the most robust common choice for chloride-containing cooling water because it is essentially immune to chloride SCC and performs well in seawater and brackish water.
My recommendation is titanium Gr 2 tubes with a compatible tubesheet and roll-joint procedure, plus design changes: increase tube velocity above about 1 m/s where erosion allows, install filtration or a strainer, improve water treatment to control scale and biological deposits, and eliminate crevices at the tube-to-tubesheet joint by using a welded tube-to-tubesheet joint or an expanded-and-sealed joint. If titanium is unavailable, superduplex with strict deposit control is the fallback, not 316L with a higher inhibitor dose.
3. Propose a monitoring and inspection plan for the replacement exchanger and explain what would trigger a shutdown.
Read the worked answer
The monitoring plan should combine water chemistry, deposit control, and direct corrosion checks. Track chloride, pH, conductivity, phosphate residual, iron and manganese, and biological counts at both inlet and outlet; monitor tube velocity and pressure drop; and inspect for deposits at the hot end. Install corrosion coupons or electrical-resistance probes in a sidestream at the same temperature and velocity, and use periodic eddy-current or IRIS inspection of the titanium tubes, focusing on the tube-to-tubesheet joints.
Shutdown triggers should include a rising iron count, a pressure-drop increase indicating deposit build-up, a coupon corrosion rate above the agreed limit, or any eddy-current indication at a roll joint. A single small leak is not an acceptable operating condition in a cooling-water exchanger, because it can propagate rapidly under deposits. The plan should also include a baseline inspection after the first 6 months and a review of the water-treatment program, since the original failure was driven by deposit-covered low-velocity areas rather than by bulk water chemistry alone.