Earlier this week the Chemical Safety Board issued a press release about their final report on the Valero Refinery Fire in Sunray, TX. This February 2007 fire resulted from a leak in a propane process line. Four personnel were injured and the refinery was out of operations for an extended period of time. Of special interest to the Chemical Facility Security community is the section in the final report that deals with near misses. The near miss of special concern was the 5,000 pounds of chlorine gas released
Chlorine Gas Release
Soon after the initial fire started it spread to a section of piping near a shed where three 2,000-lb chlorine cylinders were stored. The chlorine was used in cooling water treatment. The report describes what happened next this way:
“The fireexposed the containers to radiant heating, rupturing one (Figure 13) despite themelting of its fusible plugs, and causing the other two to vent chlorine through their melted plugs”.
By the time that this release occurred the fire crews and plant personnel had already been pulled back and moved upwind of the fire. The report notes that the 5,332 pounds of chlorine that the refinery reported released in the event could have had a toxic effect in excess of three miles from the scene of the release. The report does not note any exposure data or downwind measurement of chlorine concentration being done during or after the incident. No chlorine exposures were reported during the incident.
Chlorine Release Due to Fire
When conducting a Process Hazard Analysis (PHA) a facility should consider the ‘fire case’ when examining the safety problems associated with highly hazardous materials like chlorine. According to the CSB report (page 57) the facility did not take into account the large amounts of liquefied propane gas (LPG) in the nearby unit where the fire started. Neither did they take into account the flammable liquids in the piping passing nearby the chlorine storage location.
This failure to take into account the release of a toxic COI during a fire is repeated in the SVA process currently under way in high-risk chemical facilities. In assessing the consequences of a successful attack on an asset with a flammable COI as the primary COI the facility is required to report in the SVA the total amount of flammable COI released. There is no place to report a toxic COI or an explosive COI release as a consequence of the attack on the flammable release asset.
It can be argued that the chaotic atmospheric effects seen during a fire will make the prediction of toxic effects distance from the release problematic at best. In large fires like those experienced in this refinery blaze, the ground level winds tend to move towards the source of the fire. The out flowing air tends to move upward due to heating effects. When there is a significant directional wind in the area there is more of a horizontal component to movement of the rising air. All of this would tend to decrease the distance from the release point where there one would expect to find a toxic effect.
CSB Comments on Inherently Safer Technology
While the CSB does not comment on IST as a security measure it does mention that it is one of the principal methods for enhancing process safety. They note that:
- “In applying inherent safety principles, the preferred approach to control hazards is to eliminate them. However, if elimination is not feasible, replacing hazardous materials with less dangerous ones (substitution) should be examined”. (Emphasis added)
The CSB report does not offer any advice on what standards should be used when considering the substitution of less hazardous materials. What it does do is refer back to a book by the Center for Chemical Process Safety (CCPS); CCPS. Inherently Safer Processes, A Life Cycle Approach, AIChE, New York, 1996. This has been a recommended process safety book for chemical manufacturing facilities for some time.
The report does note that “Valero has identified replacing chlorine in cooling water treatment at all its refineries as a safety goal”. As of May of last year almost half of the Valero refineries were not using chlorine gas for their cooling water treatment systems.
The storage of toxic COI in the vicinity of flammable COI must be looked at from both a safety and security perspective. The release and ignition of a flammable release COI could result in the release of the toxic COI. This would be a complicating factor in responding to the original fire and may extend the reach of the incident to include the off-site community.
This means that facilities to look at fire protection for toxic COI storage. This includes emergency cooling systems like deluge systems. This is important when one considers that the fusible plugs on most 1-ton chlorine cylinders are designed to release chlorine at temperatures as low as 165 F. Such deluge systems would also reduce the amount of chlorine released to the atmosphere in the event of a tank venting.
Other security and safety concerns
The final report includes a number of other safety related findings that facilities might find applicable in their security reviews. These findings (page 46) include:
- The use of “ROSOVs (remotely operated shut off valves) and interlocked equipment controls to enable the safe and rapid emergency isolation of process equipment containing highly pressurized flammables”.
- A requirement to look at “deluge system activation during emergencies originating in nearby process units.”
- A requirement to look at fireproofing “for pipe rack support steel near process units containing highly pressurized flammables”.
Whether facilities address these concerns as part of their safety program or their security program is really a non-issue. Addressing these issues would be important in either case.