Chemplant Fire in Louisiana

Earlier this week there was a significant fire at a chemical facility in Duson, LA (according to the company web site the location is actually in Rayne, LA, but the news reports are consistent in calling it Duson). The facility appears to be a distribution center for the company’s oil field chemicals, though some blending may take place in at least one of the buildings on site.

The News Reports

News reports (here, here, here, and here) would seem to indicate that the fire started with an initial explosion, followed by a number (one witness reported “dozens of explosions in a half hour to one-hour period”). These reports, combined with a news photo of the incident (which shows the 10 storage tanks on site intact during the active phase of the fire) would indicate that the fire took place in the open storage of chemical containers (55-gal drums, plastic and steel totebins).

A company statement quoted here notes that “there are no uncontained spills”.

Google satellite view shows that a large portion of the facility is given over to open storage of chemical containers. Google street view of that storage area shows that it consists of a concrete pad surrounded by a 6” to 8” curb to contain spills. Street view from the back side of the plant shows that the ‘tank farm’ containing 10 storage tanks (4k-gallons each?) is on the same pad but is surrounded by an additional 8” to 10” curb.

The chemicals reported on site reported include a wide variety of oil field chemicals. Many of these chemicals would have flammability and corrosivity hazards associated with them. A close look at the Street View of the storage tanks shows that at least two of them contained flammable materials. The flammability of the typical oil field chemicals is due to the solvents used, which typically would include xylenes, methanol, toluene and benzenes.

Analysis

The investigation of this fire is ongoing and is being conducted by Louisiana authorities (the CSB typically does not investigate incidents that do not include deaths, wide scale damage, or national news focus). The results of that investigation will not be available for weeks and will probably escape news coverage unless some unusual (read ‘chargeable’) aspects are uncovered.

Based upon news storied I suspect that there was a leak in one of the storage containers on the concrete pad that contained a flammable liquid. Some sort of ignition source (static discharge?) ignited the small pool of spilled chemicals. The initial explosion reported was probably a vapor cloud near the initial fire and probably from the same spill. That small explosion would have damaged nearby containers, contributing to the spread of the fire.

The subsequent reported explosions were probably attributable to nearby containers containing flammable/combustible chemicals being heated to the extent that the gasses expanded in the container until the container was no longer able to contain the pressure. As flammable gasses exited the damaged containers, they were ignited by either the nearby fire or static discharges associated with their release.

At first glance the ‘containment’ at this facility would appear to be rather proforma. The saving grace is that the very large area encompassed by the relatively small curb results in a substantial volume of containment. I have not done the calculations, but it would seem that a large percentage of the containers on site could leak their full contents and the liquid would still be physically contained. Even if one of the storage tanks were to completely drain, it would appear unlikely that any of the contents would flow off site.

This would meet all of the legal standards for containment. The only problem would occur if the leak occurred when there was a large amount of water on site, either due to rainstorms (for which Louisiana is famous) or the application of large amounts of firefighting water. From the news reports that I have seen, it would seem that the local firefighting response did not use large volumes of water; it is usually contraindicated in flammable chemical fires in any case.

The combination of water-soluble (methanol) and water-insoluble (benzene-based chemicals) flammable chemicals on site would cause potential problems with the use of foam for firefighting as they require different types of foam to be used. Still, most fire departments are not going to have access to either type of foam. The best bet in that case is to isolate the facility and stop the fire from moving off-site.

This fire would have been more serious if one of the storage tanks had been a source of the leak that initiated the fire. A fire at the base of one of the tanks could have resulted in a weakening of the tank wall, a resulting collapse and release of the entire contents. The resulting fire and potential vapor cloud explosion would have damaged adjacent tanks spreading the fire throughout the concrete pad area. The overpressure resulting from the larger vapor cloud explosion would also probably have resulted in off-site damage (broken windows and flying debris).

One final note; as is usual with any incident at an industrial facility, at least one news report went back and looked at the regulatory history of the facility involved. They reported on the problems this facility had with the EPA’s hazardous waste regulations. While this is certainly part of the public record, these paperwork violations have little to do with chemical safety and certainly nothing to do with the current incident.