The accident on Jan. 29, 2003, killed six workers and injured 38 others, including two firefighters. The blast could be felt 25 miles away, and burning debris ignited fires in wooded areas as far as two miles away. A large fire at the plant burned for two days. The board called on North Carolina to adopt National Fire Code controls on combustible dust for industrial facilities statewide. Application of the national code, known as National Fire Protection Association (NFPA) 654, would require that businesses in all North Carolina jurisdictions adhere to recognized good practices for preventing combustible dust explosions. Those measures include segregating dust-producing operations; sealing walls, ceilings and partitions to prevent intrusion and accumulation of dust; using only electrical equipment suitable for potentially explosive atmospheres; and regularly training employees on combustible dust hazards.
The CSB report determined four root causes of the accident at West: the company's inadequate engineering assessment for combustible powders, inadequate consultation with fire safety standards, lack of appropriate review of material safety data sheets (MSDSs), and inadequate communication of dust hazards to workers.
"If the good safety practices described in the National Fire Code and elsewhere had been followed at West, this tragic accident would likely have been avoided," said CSB lead investigator Steve Selk. "We will therefore be recommending that the state of North Carolina make compliance with the dust code mandatory."
In addition to recommending that North Carolina's Building Code Council adopt NFPA 654, the report calls on the state Department of Labor to identify the industries at risk for combustible dust explosions and conduct an educational outreach program to help prevent future accidents. The report urges increased training of North Carolina fire and building code officials on combustible dust hazards. It also recommends that West improve its material safety review procedures, revise its project engineering practices, communicate with its workers about combustible dust hazards, and follow safety practices contained in NFPA 654 at all company facilities that use combustible powders.
Consistent with preliminary findings released in June 2003, CSB investigators concluded that the blast at West was caused by the ignition of a significant amount of polyethylene dust, which had accumulated above a suspended ceiling over a production area where slabs of rubber were made. The company uses the rubber to make medical items such as syringe plungers and rubber stoppers for vials. In the process, rubber strips were passed through a tank of fine polyethylene powder and water and were then air-dried with fans. Polyethylene dust with the consistency of talcum powder became airborne in the process, and the dust was drawn above the suspended ceiling by heating and air conditioning intake ducts, investigators said.
The CSB report said that while dust removal and good housekeeping were priorities at the facility, dust nevertheless accumulated above the suspended ceiling over time and went unrecognized as a serious hazard. Maintenance workers reported accumulations of one-quarter to one-half inch of dust above the tiles and other surfaces; the National Fire Code limits combustible dust accumulations to 1/32 of an inch.
"This tragic accident could have been avoided if the design and operation of the facility had taken into account the hazards of combustible dust," said
CSB Chairman Carolyn Merritt. " Unfortunately, West lacked an effective understanding of the danger, and regulations did not require adherence to control measures. Our findings underscore the need for stronger fire codes on the books and for combustible powder manufacturers to adequately identify the hazard of their products and warn customers to prevent such disasters."
CSB's investigation determined that West company files included documents warning of the explosive properties of polyethylene powder. However, the company lacked effective management procedures for incorporating those warnings into the engineering and operation of the rubber-making process. In assessing the hazards of the powder, West relied on a material safety data sheet (MSDS) prepared by a vendor that produced a polyethylene-water slurry, not the original MSDSs from the powder manufacturer, which did contain dust warnings. The MSDS from the slurry producer, Crystal Inc. PMC, did not consider the hazards once the slurry dried, a contributing cause in the accident. Among other safety recommendations, investigators urged that Crystal's MSDS be modified.
The report noted that during a previous welding operation at the Kinston plant, accumulated polyethylene powder had briefly ignited near the rubber production equipment, without causing an explosion. Although the incident indicated the combustibility of the powder, West did not conduct a documented internal investigation of this incident, which could have led to a better understanding of the dust hazard.
Selk said that extensive damage at the plant made it impossible to determine what event initiated the dust explosion but noted the report identifies four theories: a batch of rubber that overheated and ignited; an electrical ballast or light fixture that ignited accumulated dust; a spark caused by a possible electrical fault; or ignition of dust in a cooling air duct feeding an electric motor.
"The ultimate source of the large explosion that destroyed the plant was the dust accumulation, and that is where future efforts need to focus. Without accumulated fuel, dust explosions simply do not occur," said Selk.
The explosion in Kinston was one of three fatal dust explosions in 2003 under investigation by the CSB. A phenolic resin dust explosion at an automotive insulation maker in Corbin, Ky., in February caused seven fatalities and injured 42, and an aluminum dust explosion at an Indiana automotive parts maker in October killed one worker and burned two others. These and other recent dust explosions are a focus of CSB's investigation on combustible dust hazards, which is currently underway. That study will review possible national initiatives to reduce the occurrence of industrial dust explosions.