CSB Interim Report: Catastrophic Failure of a Kettle Reboiler Shell
The Chemical Safety Board has just released an update to its ongoing investigation of the explosion of a kettle reboiler at the Honeywell Performance Materials facility in Geismar, Louisiana. Large quantities of hydrogen fluoride (HF) and chlorine were released. There were no injuries.
Salient facts include the following:
The carbon steel shell of the reboiler was subject to uniform corrosion.
The reboiler had been designed to meet ASME code.
The report does not tell us what the operating pressure in the shell was. However, the Maximum Allowable Working Pressure (MAWP) was 200 psig (14 bars).
The reboiler shell had already been replaced once as a result of corrosion. It was due for another replacement at the next opportunity.
The explosion was a BLEVE (Boiling Liquid Expanding Vapor Explosion). The image shows the total destruction of the reboiler shell. (The tube bundle looks to be in good condition.)
Metallurgical testing following the explosion showed that parts of the shell were just 0.03 inches think ― the minimum allowable thickness was in the 0.08 to 0.16 inch range. (The obvious question to ask is, ‘Should the reboiler shell thickness be tested more frequently?’)
There were no workers present at the location at the time of incident.
Additional questions worth considering are:
Would a thermosyphon reboiler be a safer choice?
Would it be possible to have the bundle inside the column?
What is the steam pressure in the bundle, and what would be the effect of a tube failure?
Had there been a recent change to process conditions (temperature, pressure, liquid composition) that may have accelerated the rate of corrosion?
Are there emergency isolation valves on the reboiler feed and discharge lines?
Could instrumentation check the metal thickness on a continuous basis?
Are the other sections of the column safe, including instrument connections, trays, relief valves?
The investigation is still under way, so we will see what the final report says.
