33 hours, 3 Toxic Releases, 1 Fatality: All preventable, says Chemical Safety Board of January 2010 incidents at DuPont Belle, West Virginia plant

On the afternoon of Saturday, January 23, 2010, Carl "Danny" Fish, a 32-year employee of the DuPont plant in Belle, West Virginia was performing a routine operation when a hose carrying phosgene (a chemical so toxic it was used as a weapon during World War I) ruptured, spraying him in the face and chest. Fish was rushed to the hospital. He died the night of January 24. Two workers who attended to Mr. Fish were also exposed to phosgene but apparently without any lasting impact.

What initially sounded like a freak accident turns out to have been but one in a series of equipment failures that, according to the U.S. Chemical Safety Board (CSB) draft investigation report released July 7th, resulted from a combination of poor maintenance, inadequate provisions for handling toxic materials, and faulty alarm and hazard alert systems. The fatal phosgene release was the third accidental release of a highly toxic substance to occur at the Belle DuPont plant within the space of 33 hours. It happened while the plant was on a "safety pause" following a "near miss" event involving a frayed phosgene hose of the same kind that broke just hours later, spraying the chemical that killed Danny Fish. Documents included in the CSB report show that DuPont had considered but rejected as too costly measures that could have prevented this incident.

These hoses - made from a material subject to corrosion - were supposed to be changed monthly, but the one that broke had not been replaced in seven months. Poor inspection procedures, inadequate alarm systems, and faulty safeguards allowed leaks of hazardous chemicals to develop and persist unnoticed. Communication about the accident and hazards involved was inadequate and left potentially significant gaps in what plant employees, emergency responders, and the surrounding community needed to know to be as safe as possible from potential toxic chemical exposures.

33 hours - three toxic releases
Phosgene wasn't the only chemical released by the Belle DuPont plant on January 22nd and 23rd, 2010. Methyl chloride - a potent cardiovascular, liver, and neuro-toxin - and oleum, (a highly corrosive form of sulfuric acid; the oleum released had an acid content 20 percent greater than sulfuric acid), also leaked from improperly maintained equipment on the following timeline:

• Friday, January 22, 2010: A leak of methyl chloride that had apparently gone undetected for five days was discovered at the Belle plant. By the time it was stopped, approximately 2,000 pounds of the chemical had been released into the atmosphere.
• Morning of Saturday, January 23: DuPont workers reported a "fuming cloud" above a chemical delivery pipe. About one-half hour later, valves were closed to stop a leak of oleum, some 22 pounds of which had escaped through a hole in a corroded pipe.
• Afternoon of Saturday, January 23: A phosgene transfer hose broke and sprayed Danny Fish with enough of the chemical to cause his death.

What stands out from the CSB report is that all of the investigated incidents were entirely preventable and that DuPont had failed to take corrective action when hazardous conditions that eventually led to these toxic chemical releases were identified. Each "incident was preceded by an event or multiple events that triggered internal incident investigations by DuPont, which investigated all of these precursor events and issued recommendations and corrective actions," writes the CSB. "Despite investigating these preceding events the recommendations and corrective actions did not prevent the occurrence of similar events."

In the year following the three incidents under investigation, the CSB recounts that the Belle plant had another methyl chloride release in September 2010 during which more than 160,000 pounds of methanol were estimated to have been released into the nearby Kanawha River in the space of 24 hours. Then in December 2010, two workers received first-degree chemical burns to their faces while taking samples of monomethylamine during what should have been a routine operation. Both the September and December incidents at the Belle plant occurred after an EPA inspection prompted by the January incident resulted in the EPA ordering DuPont to undertake a safety review, and issuing a statement in March 2010 saying the facility failed to comply with Clean Air Act requirements designed to prevent accidental releases of hazardous chemicals.

(The Belle plant was not the only Dupont with safety problems in 2010. In November 2010 an explosion in a polyvinyl fluoride tank at the DuPont plant in Tonawanda, NY (near Buffalo) killed a worker who'd been hired to clean the tank and injured one of his colleagues.)

Underscoring the gravity of its assessment is the CSB characterization of DuPont as a company that had been considered an industry leader in health and safety. "According to company documents, the plant had the best safety record of any DuPont production facility prior to the incidents of January 22 and 23," writes the CSB. "We were surprised and alarmed" to learn of these incidents, said CSB Chair Rafael Moure-Eraso during a July 7 press conference.

Deciding against investments in safety
But none of these incidents should have surprised DuPont, according to information detailed by the CSB. The CSB report chronicles decisions made by DuPont in its choices of equipment and safety measures for toxic materials handling at the Belle plant, particularly for phosgene, that rendered plant operations more hazardous - more hazardous than those of at least one other DuPont plant. But there had been issues with other chemical delivery systems at the plant. A similar oleum leak had occurred at the plant in 2009, and alarm systems for that chemical process and for methyl chloride releases were chronically faulty.

In 1988, according to documents in the CSB report, DuPont had considered an enclosure and scrubbers for its phosgene storage but decided against these measures after a cost-benefit analysis. DuPont had also rejected using a stronger hosing material, less vulnerable to chemical degradation and had failed to thoroughly inspect existing hoses for what turns out to have been chronic corrosion. (These hoses were fraying underneath a plastic tag, and although the material was not ideal, changing hoses monthly as schedule required would have helped avert the breaks.)

"It may be that in the present circumstances the business can afford $2 MM for an enclosure however, in the long run can we afford to take such action which has such a small impact on safety and yet sets a precedent for all highly toxic material activites [sic]," says a 1988 DuPont memo (included in the CSB report) discussing the pros and cons of enclosing its Belle phosgene shed. However, DuPont's phosgene operation in Mobile, Alabama is enclosed, which prompts the question: Why weren't comparable safeguards established in West Virginia?

At Belle, workers entering the phosgene shed for duties like those Danny Fish performed were not required to wear respiratory protection or chemical-resistant coveralls. In 2004 a DuPont process hazard analysis at Belle, undertaken by the company after previous phosgene release incident, prompted a recommendation to enclose the phosgene shed. The deadline for enclosing the shed was extended four times and remained unfulfilled at the time of the fatal accident in 2010. "The shed contains no mechanical ventilation or exhaust systems to control phosgene leaks, only natural ventilation flowing through the shed wall opening from the atmosphere," writes the CSB.

The CSB investigation also highlights the fact that the Belle DuPont plant does not maintain the same safety measures for handling phosgene as recommended by the National Fire Protection Association. It also notes that OSHA's current standards for phosgene handling and storage are 1965 standards. According to the CSB report, in a safety audit of the plant conducted in 2009, OSHA did not address consequences of failures in preventative maintenance of phosgene hoses or hose problems due to corrosion or thermal expansion of the gas - both of which occurred in the hoses that broke.

Updating OSHA standards for phosgene handling, improving safety for phosgene operations at the Belle plant, and upgrading alarm systems for other toxic materials systems, including methyl chloride and oleum, are among the CSB report recommendations. But other changes in operations may render some of these recommendations moot.

As a result of an unrelated earlier EPA inspection, in 2009 EPA ordered the Belle DuPont plant to upgrade emissions monitoring and abatement in the unit where the oleum was released. As part of a consent decree issued by the EPA, that unit of the plant, owned by Lucite International but run by DuPont employees, was shut down in March 2010. Earlier this year, DuPont idled the unit where phosgene was used as a process chemical for pesticides, according to the company, due to changes in market conditions.

Methyl chloride, oleum, and phosgene have all been used at the Belle plant as chemical intermediates in the manufacture of agricultural chemicals, polymers, and other chemical products. While existing safety measures for handling such toxics are improved, the most fundamental question would seem to be: Can companies like DuPont continue to produce plastics, pesticides, and other needed synthetics but do so without relying on dangerous chemicals that pose such hazards to workers and communities where these plants are located?

Elizabeth Grossman is the author of Chasing Molecules: Poisonous Products, Human Health, and the Promise of Green Chemistry, High Tech Trash: Digital Devices, Hidden Toxics, and Human Health, and other books. Her work has appeared in a variety of publications including Scientific American, Salon, The Washington Post, The Nation, Mother Jones, Grist, and the Huffington Post. Chasing Molecules was chosen by Booklist as one of the Top 10 Science & Technology Books of 2009 and won a 2010 Gold Nautilus Award for investigative journalism.