Sampling the Plastimet Fire

Laurie Allan, a Chemistry graduate student, and I had just finished setting up an air sampler on the roof of the company across the street from the Plastimet Fire, Hamilton's worst industrial fire. The weather was still hot and humid in the late afternoon of July 11, 1997, just hours after 650 residents near the fire had been evacuated from their homes.

Our respirators allowed little of the acrid smell of the fire to reach our lungs. At times we had a clear view of the fire-fighting activities only one hundred metres away; seconds later we were completely engulfed in smoke. By the time the fire was finally out, 200 tonnes of poly(vinyl chloride) (PVC) and 200 tonnes of other plastics, primarily polyurethanes, had been consumed. Our analyses would show the smoke contained very high levels of particulate material, aromatic hydrocarbons, dioxins and furans, phthalates and other organic compounds. Once the fire had been declared out, the levels of these contaminants dropped dramatically to normal urban values.

The above photo, showing the view due south across Hamilton Harbour from Burlington, was taken at 8 p.m. on July 9, about 15 minutes after the fire had begun. The plume from this conflagration was visible from as far east as Toronto, minutes after it started under mysterious circumstances at Plastimet, Inc., a plastics recycling plant. This fire made headlines across Canada and would capture the attention of the local media and Hamilton-Wentworth residents for the next several months.

There were widespread concerns over potential health effects of chemical emissions from the fire, concerns over dioxins settling on vegetables and concerns about impacts of the thousands of litres of fire water that might carry chemical contaminants into the harbour or ground water. Greenpeace representatives arrived in Hamilton soon after the fire warning residents of the dangers of dioxins and furans that are released from PVC fires. Ultimately, a class action lawsuit was filed on behalf of local residents, fire fighters and some local businesses.

As information about Plastimet's operations were unearthed during the days and weeks following the fire, questions about the cause of this fire and why the company was even allowed to continue to operate at the site went unanswered. Local residents and Ministry of Environment officials were all too familiar with problems of metal contamination at the site as a result of the operations of a company called Usarco which had gone into receivership in 1990. It seems that Plastimet Inc. had set up operations in this abandoned building in 1995 without bothering to get a business licence from the City of Hamilton.

Of immediate concern to firefighters and local residents during the fire was information about the nature and hazards posed by the chemicals released from plastics fires such as this one. While there is some documentation of emissions from small scale plastics combustion under controlled conditions, surprisingly little is known about the chemical impacts of large scale plastics fires. Our research group has had experience collecting and analyzing ambient air samples in Hamilton and Toronto and near the 1990 Hagersville Tire Fire. We saw an opportunity to apply our sampling and analytical expertise to the identification of the chemicals released from this plastics fire and to assess the potential long term health effects of these exposures on residents living close to the fire.

Ministry of Environment (MOE) officials graciously allowed us to set up an air sampler in the area surrounding the fire. We searched near the fire site and settled upon a location on a roof across the street from the fire. We hauled our sampler up to the roof and located it overlooking Wellington Street and Simcoe Streets, about 100 metres from the fire. The building was without power but the company kindly allowed us to use one of their gasoline-powered electricity generators and 150 feet of power cable to supply power for our sampler. As grey-brown smoke wafted toward us 10 metres above Wellington Street, we knew that we had found a good location downwind of the fire.

The first air particulate filter was collected between 5:30 and 10:30 p.m. while a second one was collected between 10:30 p.m. and 3:00 a.m. the next morning. Both of these filters had collected jet black material. A third filter collected the next morning immediately after the fire had been declared out by the Fire Chief was pale grey in colour. We took these filters back to our lab at McMaster for weighing, extraction and chemical analyses.

The particulate loadings on the three filters were 1000, 590 and 60 mg/m³, respectively; the current MOE guideline for inhalable particulate is 50 mg/m³. At particulate levels above 100 mg/m³, many people, particularly children, the elderly and asthmatics, complain of breathing difficulties. We estimated that outdoor concentrations of particulate about 0.5 km from the fire were in the 100-200 mg/m³ range during the latter stages of the fire; indoor concentrations of particulate would have been much lower had residents kept their doors and windows shut. Once the fire was out, inhalable particulate levels quickly returned to values that are typical in that area of Hamilton.

The filters were then extracted with organic solvents and the levels of aromatic hydrocarbons, dioxins and furans and other compounds determined using gas chromatography-mass spectrometry. During the fire the levels of aromatic hydrocarbons at our sampling site were 5000 ng/m³, a value 2000 times greater than a typical urban background level. Residents standing outdoors 0.5 km from the fire would have been exposed to very high aromatic hydrocarbon levels, about 5- to 10-fold lower than measured at our sampling site. However, immediately after the fire was out, these levels dropped to 8 ng/m³, a 600-fold decrease, a value close to the annual average for downtown Hamilton.

The MOE dioxin lab in Toronto kindly agreed to perform a dioxin/furan analysis on the sample collected during the fire. The value they reported to us was the highest value their lab had ever determined in an air sample - 1000 pg/m³ TEQ , a value 2,500-25,000 times higher than ambient levels determined for Hamilton air a few years ago. We estimated the dioxin concentrations at various distances from the fire, calculated the exposures of local residents due to inhalation and estimated their risk due to these exposures.

In the worst case scenario, a resident who remained outdoors for 24 hours in the smoke plume only 300 metres from the fire would have received a dioxin/furan exposure in one day equivalent to what they would have normally received over three and a half weeks due to their dietary intake of these compounds. More realistically a resident who stood outdoors in the plume for 4 hours and remained indoors for 20 hours would have received a 24 hour exposure equivalent to what they would normally receive in a week. These dioxin exposures were short-term exposures and do not pose an unacceptable long-term health risk to residents. The effects from exposure to aromatic hydrocarbons also do not pose a health threat.

Some local residents suffered short-term health problems resulting from inhalation of particulate material, carbon monoxide, hydrogen chloride and some volatile organic compounds; these effects would have been manifest almost immediately, particularly in sensitive individuals, but would have disappeared within a week or two. The incidence of hospital admissions and visits to family doctors by local residents during and after the fire was slightly greater than normal.

In summary, our samples provided us with an opportunity to determine the chemicals released from this fire and to predict the concentrations of these substances downwind of the fire. We were able to assess the potential impacts due to exposure to these substances using a risk-based approach.

Brian McCarry