Problem No. 78 HP stripper leakage – How should it be managed?

Any leak in high-pressure urea equipment is critical, as corrosive ammonium carbamate can attack the carbon steel pressure-bearing parts – including the vessel wall and tubesheets – and may ultimately lead to catastrophic rupture. The image shows corrosion of a carbon steel tubesheet in a falling film high-pressure carbamate condenser caused by a leak at the tube-to-tubesheet weld.

Shutting down the plant in the event of a leak appears to be the most advisable option. But what if the leak is so small that it cannot be located even after shutdown? This Round Table discussion explores that dilemma and offers valuable suggestions based on many years of experience.

Zavtra Nguyen from PetroVietnam, Vietnam starts off the discussion: Can you share any experience in dealing with HP stripper leakage? Have you encountered different types of leaks? The most common are tube-to-tubesheet defects, is that correct? How did you detect the leak point and how did you address it? Which expert teams specialise in handling these issues? Please share your cases and experience. Thank you.

Mark Brouwer of UreaKnowHow.com in the Netherlands joins the discussion and asks for some clarifications: What materials of construction does your stripper have? How old is the HP stripper?

Zavtra replies: The stripper is bimetallic: Zr/25-22-2; age: 13 years.

Mark provides some initial comments and suggestions: Bimetallic strippers in Saipem plants often fail at the tube ends on the bottom side, where temperature – and therefore corrosion rate – is highest. Any leakage should be detected by an increase in the conductivity of the steam condensate.

After opening the equipment, an ammonia leak test is preferred over helium testing, because ammonia can penetrate moisture present in the leak path. Sufficient time should be allowed for the ammonia to pass through the leak path. You can contact Saipem or AXO Welding who perform the repairs.

Zavtra asks another question: Does anyone have experience operating an HP stripper that is leaking, and what control measures are recommended – especially regarding the plant steam system and related equipment? When should we stop the plant to address the leak?

The steam condensate conductivity has increased from 9 to 18 µS/cm and is now stabilising. We are operating the plant at a lower temperature by about 0.5°C, and seeking experience on whether the plant can continue running until a major maintenance shutdown. Any recommendations?

Mark comments further: Leaking carbamate from a tube-totubesheet connection can corrode the carbon steel tubesheet and threaten vessel integrity. Experience from high-pressure falling film carbamate condensers shows that significant portions of the carbon steel tubesheet around a leaking tube can be corroded away relatively quickly.

In these condensers, 4–5 bar steam is formed on the shell side from circulating steam condensate. In a high-pressure stripper, however, 20 bar steam condenses into steam condensate on the shell side, so one could argue that leaking carbamate is more diluted than in a falling film condenser.

Eddy current testing (Permascope) can be used to inspect the condition of the carbon steel tubesheet around the tubes. JSC Inspections has relevant experience in this.

Has anyone done this in a stripper? Has anyone inspected the condition of the tubesheet of a stripper that was no longer in operation? This would be valuable information regarding the risk of corrosion in the carbon steel tubesheet.

Prem Baboo, retired from Dangote Fertilizers, Nigeria shares his experience: My recommendation for operating a urea bimetallic stripper is:

• Stripper bottom temperature should be 204°C
• Passivation air should be sufficient at the stripper bottom
• The N/C ratio in the reactor should be 3.5

Rama Raghava Kumar Kotti, Freelance Plant Operations Consultant, India joins the discussion: Here are some guidelines and considerations based on my experience in urea plant operations:

Leak confirmation: How was the leak confirmed, given that the conductivity increase from 9 to 18 µS/cm is relatively modest?

Ensure the change is not due to contamination from steam condensate, flushing condensate, reverse flow in flushing networks, or direct injections.

Sampling and monitoring: Have samples been taken from the MP steam at the stripper inlet and the condensate outlet of the stripper? What values were recorded?

At a minimum, monitor the following points: stripper outlet condensate, carbamate condenser outlet, steam condensate tank, LP steam network.

Contamination checks: If there is low-pressure steam injection in the CO2 turbine, check for contamination and stop the injection if necessary to prevent plant-wide contamination.

Conductivity monitoring: How frequently is condensate conductivity at the stripper outlet being monitored? Continuous monitoring is essential until the leak is addressed. Do not assume the current trend will remain stable until the major maintenance shutdown.

Frequency of monitoring: Once a leak is confirmed, increase monitoring frequency to at least once per shift.

Trend analysis: Plot continuous trends for these analyses until major maintenance is carried out, provided the leak does not suddenly worsen.

Plant shutdown and emergency instructions: Exercise caution and ensure thorough, continuous monitoring, as operating with a leak is risky. The decision on when to stop the plant depends on the rate of leak progression and is typically made by top management.

The leak may worsen gradually or suddenly, so operating personnel should have clear instructions on when to shut down before the situation becomes an emergency, with defined operating limits.

Technology suppliers’ recommendations should be considered, especially for major equipment such as the HP urea stripper.

Bob Edmondson, retired Yara Canada joins the discussion and adds a note of caution: Although the concentrations of CO2 and NH3 in steam condensate sampled from the steam saturator may be relatively low in the case of a small leak, the concentrations are much higher on the shell side of the stripper itself. The stripper shell operates as a rectifying section, while the saturator acts as a highly effective stripping section.

I suspect that corrosivity is only a factor at the bottom of the stripper shell, since condensation higher up occurs only on the tubes, which have good corrosion resistance.

Another consideration is that a leaking synloop boiler in an ammonia plant can produce the same symptoms in terms of condensate conductivity. Distinguishing between the two requires monitoring turbine condensate conductivity without the cation cartridge in the ammonia plant.

Majid Mohammadian, Fertiglobe, UAE shares his experiences and asks: Do you have a conductivity analyser in the stripper steam saturator? To confirm whether the source is a leak in the stripper tube or tubesheet, I recommend installing an analyser in the saturator drum and continuously monitoring changes.

Rama provides follow-up comments and responds with additional clarification:

Contamination levels and plant operation: I appreciate your concern regarding the higher CO2 and NH3 levels on the shell side of the stripper, but the current situation involves contamination at ppm levels, with ammonia concentrations ranging from 5 to 100 ppm, or slightly higher. These levels are far below the percentages being referenced, which would require immediate shutdown.

At such high contamination levels, the plant would not be operable because steam and condensate purity would be severely affected. Therefore, the current operational guidelines and monitoring are based on lower contamination levels where the plant remains in service.

Stripper bottom corrosion: Given the lower contamination levels, the corrosion risk at the bottom of the stripper shell is considered minimal. At ppm concentrations, corrosion rates are expected to be negligible, which may allow the plant to continue operating until an immediate or opportunity shutdown becomes available.

This distinction is important because the corrosion concerns raised are more relevant at much higher contamination levels.

Decision-making for shutdown: If the leak is minor and a shutdown has already been taken, it may still be difficult to identify the leak source, even in a synthesis loop boiler, potentially resulting in production loss. For that reason, the shutdown decision should be made by the technical team only after thoroughly analysing and confirming the data.

Sampling and monitoring protocol: I have suggested taking samples at the immediate inlet and outlet of the stripper using a small condenser, while ensuring that ammonia and other gases do not escape during sampling.

The main objective is to monitor the increasing trend of ammonia and CO2 levels, even if small sampling errors are unavoidable. The focus should remain on detecting and analysing trends to support decision-making.

Field verification of the stripper: There are typically vent or drain points at the immediate inlet and outlet of the stripper. These should be checked in the field to ensure that sampling and monitoring are accurate. Proper field verification and adherence to sampling procedures are essential for obtaining reliable data.