A large explosion and fire that took the lives of two workers at the Bayer CropScience (Bayer) plant last August was caused by a thermal runaway reaction during the production of an insecticide, according to a U.S. Chemical Safety Board press release. The event likely resulted from significant lapses in chemical process safety management at the plant, CSB investigators said in preliminary findings released at a news conference.
The blast on August 28, 2008 in Institute, West Virginia occurred as the runaway reaction created extremely high heat and pressure in a vessel known as a residue treater, which ruptured and flew about 50-feet through the air, demolishing process equipment, twisting steel beams, and breaking pipes and conduits. Two operators died as a result.
Eight workers reported symptoms of chemical exposure, including aches and intestinal and respiratory distress, including two employees of the Norfolk Southern railway company and five Tyler Mountain, West Virginia volunteer firefighters, and an Institute, West Virginia volunteer firefighter. Two sought treatment at a hospital emergency room the next day, were treated, and released.
“The explosion occurred within 80 feet of a pressure vessel containing more than 13,000 pounds of methyl isocyanate, or MIC, a raw material for the pesticide the company was making at the time, and the same chemical that caused death and injury in the Bhopal accident 25 years ago,” CSB Chairman John Bresland said.
He added: “As our investigation continues, we will look further into the issues surrounding the safe placement of the tank and its potential vulnerability. We note that other chemical companies, notably DuPont, no longer store MIC in their chemical production and we are looking into other systems that make and then immediately use the MIC, eliminating the need for storage.”
Bayer CropScience in Institute is a large chemical complex of more than 400 acres that was first constructed in the 1940s. Until 1986, it was owned by Union Carbide which produced carbamate pesticides at the site. It was acquired by Bayer in 2002, and now has more than 500 employees.
CSB Lead Investigator John Vorderbrueggen said, “Prior to starting up, Bayer had recently upgraded the computer control system for the unit, replacing a Honeywell system with one purchased from Siemens. The control screens were completely different - and Methomyl production equipment control was changed from a keyboard to a computer mouse - yet operators had not been fully trained and prepared to operate the complex process equipment on the new system. Furthermore, the written operating procedures for the unit were significantly out of date and did not adequately address all process equipment startup and normal operating steps.”
Vorderbrueggen said that the residue treater, a large pressure vessel, had an undersized heater. “According to unit operators, the heater for the residue treater was incapable of reaching the required temperature to begin the controlled decomposition of Methomyl.”
As a result of the longstanding heater problem, operators had to use a workaround. This involved defeating safety interlocks controlling flow into the residue treater vessel. The CSB found a normalized practice outside of operating procedures of starting to feed Methomyl into the vessel below the required temperature in order to create the necessary heat for the startup. But bypassing the interlocks made it more likely that too much Methomyl would enter the vessel. Safety analyses, and the operating procedure warned that Methomyl concentration above one percent inside the residue treater would likely cause it to violently rupture.
“As a result of equipment deficiencies, improper procedures and lack of training on brand-new computerized control equipment,” Vorderbrueggen said, “the vessel was charged with as much as a 20 percent solution of Methomyl in solvent, whereas the residue treater was designed to safely decompose the chemical at a concentration of less than one percent in solution.”
The CSB reported that operators attempted to check the residue treater vent system as the pressure rose. But the residue treater ruptured, suddenly released 2,500 gallons of Methomyl-solvent liquid and chemical decomposition products.
“These equipment deficiencies and procedural deviations were never subjected to formal management-of-change reviews to assess their safety - a key requirement of the OSHA process safety management (PSM) standard,” Chairman Bresland said.
The CSB is examining operator fatigue as a possible contributor to the accident. Unit operators worked very high overtime levels during the three months prior to the accident, averaging almost 20 hours a week of overtime. Operators repeatedly worked 12-hour days, and sometimes up to 18 hours, with very few days off. Chairman Bresland said, “We are concerned about the potential for operator fatigue, which can of course be an important factor in major accidents.”
Bresland said it was fortunate that pieces of the ruptured vessel did not impact the pressure vessel containing highly toxic methyl isocyanate, known as MIC, which was located approximately 80 feet from the explosion site. “An impact on this tank could have the potential of causing a release of MIC which could cause injury or death to those exposed,” Bresland said. The tank is surrounded by a large wire-rope protective mesh shield designed to prevent impact. He said the CSB plans to study the design basis for the protective shield to determine whether the MIC tank is appropriately located and protected.
In addition to a discussion of the chemical process involved and the timeline of events leading to the accident last August, the CSB community briefing this evening will include a review of emergency response communications problems, chemical exposure symptoms for Methomyl and methyl isocyanate (MIC), and placement and protection of the MIC storage vessel. Chairman Bresland noted the CSB investigation is continuing, and that safety recommendations will be issued in the final report, expected later this year.
DÃ©jÃ vu: Did Institute, W. Va., narrowly escape a Bhopal-type catastrophe? (4/24)
April 24, 2009