Chloride Cracking in a 304L Heat Exchanger
A 304L stainless steel heat-exchanger shell operates at 60°C with cooling water containing 150 ppm chloride. The outside is insulated and occasionally wetted by washdown. After 8 months, dye penetrant shows branched cracks on the shell near several nozzle welds. Metallography shows transgranular, branching cracks with no carbide precipitation at grain boundaries. An ASTM A262 Practice E test on a field sample passes. A second, identical unit is made of 316L and operates at 45°C with 80 ppm chloride; it has no cracks so far. Operations wants to replace the failed shell with 316L and continue at the same temperature and chloride level. Maintenance argues that because the A262 test passed, the 304L was not sensitized and therefore the cracking must be a welding defect.
1. Identify the most probable failure mechanism and explain how the crack morphology and the A262 pass result support it.
Read the worked answer
The mechanism is chloride stress-corrosion cracking (SCC), not sensitization and not a welding defect. The evidence fits SCC:
- Transgranular branched cracks are typical of chloride SCC in austenitic stainless steel.
- Cracking is concentrated near welds, where residual tensile stress is high.
- The service temperature of 60°C with 150 ppm chloride is within the practical SCC range for 304/304L; the outside insulation can allow chloride to concentrate by evaporation.
- An A262 Practice E pass shows that the material is not sensitized and would resist intergranular corrosion. It does not test resistance to chloride SCC. The low carbon content of 304L improves weld sensitization resistance but does not make the alloy immune to chloride SCC.
A welding defect would not normally produce branched transgranular cracks running through the parent metal away from the fusion line. The holiday in the insulation and the washdown water are the likely chloride source.
2. Evaluate whether replacing 304L with 316L is sufficient for the same service (60°C, 150 ppm Cl-) and for the second unit (45°C, 80 ppm Cl-).
Read the worked answer
316L is not a reliable fix for the failed service. Molybdenum improves chloride SCC resistance, but 316L is still an austenitic stainless steel and is not immune. At 60°C and 150 ppm chloride, with wet insulation and weld residual stress, 316L is at or beyond its practical threshold and would be expected to crack eventually.
For the second unit at 45°C and 80 ppm chloride, 316L is more likely to survive, but the margin is not large. Evaporation under insulation can concentrate chloride well above 80 ppm, so the bulk-water number understates the local environment. If the insulation stays dry and the temperature stays at 45°C, 316L may be acceptable; if wetting is possible, it is a marginal choice.
A safer material for 150 ppm at 60°C is a duplex stainless steel such as 2205, or a 6% Mo superaustenitic alloy such as 254 SMO or AL-6XN. The final choice should be based on a chloride SCC threshold chart, not on the generic ranking of 304 versus 316.
3. Recommend three changes (material, design or process, and operation) and state what you would inspect or test to confirm the diagnosis.
Read the worked answer
Material: for the 60°C, 150 ppm service, specify duplex 2205 or a 6% Mo austenitic alloy. Confirm by a coupon or slow-strain-rate test in the actual water, not just by alloy data sheets. For the 45°C, 80 ppm unit, keep 316L only if the insulation can be kept dry and chloride concentration is controlled.
Design/process: eliminate chloride traps. Use closed-cell insulation, vapor barriers, and good drainage; avoid crevices at nozzles. Specify a weld procedure that minimizes residual stress, and stress-relieve where practical. Do not rely on passivation alone; it does not prevent SCC.
Operation: limit chloride to a level consistent with the selected alloy, keep the external surface dry, and monitor washdown water quality. If temperature can be lowered, do so; even 10°C can matter.
Inspection and confirmation: PT or ACFM of all welds and wet insulation zones, UT for crack depth, and metallographic replication. Confirm the chloride source by deposit analysis. Hydrotest is not a reliable SCC test. For the replacement, require a documented SCC resistance test in the actual environment.