Fire Hazard or Fatal Risk

I am learning a lot these days thanks to discerning readers like Scott Anderson of URS Corp. Thank you, Scott. I will be more vigilant in reviewing material from EPA and the Department of Justice regarding the risks of anhydrous ammonia and other hazardous gases.

Anderson wrote to me:

I can not help but comment on "Hershey Creamery Fined in First RMP Prosecution" in a recent Environmental Protection e-newsletter as it is misleading in one of the key facts presented.

The article says: "Anhydrous ammonia is regulated under the CAA because it is flammable and can irritate both the skin and eyes."

I disagree. The risk of anhydrous ammonia is not based on its flammability. The risk is that it is a toxic gas that can kill people! This is the reason it is regulated by CAA 40 CFR 112(r) and a Risk Management Plan is required if your process stores over 10,000 pounds of it.

The risk is also not trivial: a "worst case release scenario" of losing 5,000 pounds over a 10 minute period, in stable, calm wind conditions could adversely affect everyone within a 0.9 mile radius!! (See Example 2 in "Supplemental Risk Management Program [RMP] Guidance for Ammonia Refrigeration Facilities" published by EPA, link: http://www.epa.gov/emergencies/docs/chem/Appendix-E-final.pdf.)

In one source, I found this summary regarding flammability: Anhydrous ammonia is classified by the DOT as a non-flammable gas. Ammonia vapor is flammable only over a narrow range of 16 percent to 25 percent by volume in air and a strong ignition source must be present.

The RMP "toxic endpoint" for ammonia is 200 ppm (0.14 mg/L). This airborne concentration has been published by the American Industrial Hygiene Association and is the maximum airborne concentration below which it is believed that nearly all individuals can be exposed for up to one hour without experiencing or developing irreversible or other serious health effects or symptoms which could impair an individual's ability to take protective action.

What other incorrectly described risks have you encountered in publication? It might be helpful to develop a list and generate a more solid knowledge base in the field. What do you think?

Posted by L.K. Williams, EPonline on Dec 08, 2008