Microbiologically induced corrosion in firewater system

Client challenge

Our client used rainwater collected in basin and sumps as firewater (FW) in its main process units. Integrity of its FW system was under question as susceptibility by microbiological induced corrosion (MIC) was expected and the risk of failure due to the same. The client noted absence of a water treatment program and trends in water analysis were not available, however general wall loss was noted in historic inspections and no major leaks ever recorded in the operation history of 10 years in the plant.

How we helped

Vysus Group recommended water sample analysis be taken from three locations, from basin to piping segments after review of piping and instrumentation diagrams (P&IDs). Sulfate reducing bacteria (SRB) and acid producing bacteria (APB) counts were required for review from the analysis by an independent laboratory. Dead leg locations were targeted for water sample analysis as they are the most susceptible locations for MIC.

Review of the provided inspection summaries was performed to understand issues reported related to corrosion and/or leaks. UT thickness analysis at several locations was performed to understand remaining life based on corrosion rates that was obtained.

Powerful results

The inspections review showed pitting as the most common form of corrosion. Leaks that may have happened in the past were not recorded so the root cause for those incidents were not known. Thickness for several circuits showed remaining life greater than 10 years from the present date except for small bore dead legs which showed less than 5 years for a few circuits. Water analysis from the three locations selected did show presence of SRB and APB.

The upstream industry has conducted several researches on SRB and APB counts where thresholds are mentioned in publications. Vysus noted counts were an order of two magnitude less than those researched counts in the bulk analysis that was performed. But colonies of the bacteria when present will raise the counts exponentially at the MIC attack locations. The bulk counts variation at regular intervals by 10-15% from established trends is at times an indicator of MIC in the system.

Flushing and draining of the system coupled with bacteria scavengers was recommended to the client to avoid any noticeable fluctuations in the trends.