Forecasted population growth in rural/urban interface areas will expose more residents and homes to an increased bushfire risk. As a result, fighting bushfires at the rural/urban interface is likely to become more frequent, but currently little is known about the air toxins emitted and exposure concentrations inhaled by firefighters and residents.
The rural/urban interface is characterised by multiple fuel types, including vegetation, as well as a range of combustible materials from house structures, house contents, vehicles and other objects around a house. These burning materials are likely to emit additional toxic combustion products, and as a result may cause a greater health risk to firefighters and community members in the vicinity of the fire.
Tests on a small scale have been conducted on a range of materials present in buildings to assess the type and yield of a range of emission products. Furthermore, large scale fire tests have been conducted to assess emissions from car fires and fires within a specific room. A review of the existing literature has revealed that organic compounds were considered to present a potential health hazard, but in most studies individual compounds have either not been identified or quantified. Although total emissions of carbon monoxide (CO) and carbon dioxide (CO2) dominate, hydrocarbons and volatile organic compounds are important contributors to the total emissions, and therefore likely to impact on health.
What Research Has Been Done?
Experimental burns were conducted using a cone calorimeter, which is a small-scale fire test apparatus that aims to determine heat release rate, mass loss rate, ignitability, gas species and particle production rate. Specimens sized 100mm by 100mm of various building types and furnishings were conditioned at 23°C and 50 percent relative humidity. The specimens were then placed into a sample holder and subjected to a radiant heat source of 25 kW/m2. The tests were conducted in well-ventilated conditions with constant air supply fed into the system. The gases passed through an exhaust duct, which was lined with two sampling inlets. These were linked to a number of sampling devices to collect a range of gaseous species and particles.
A desktop study on toxic emissions from fires at the rural/urban interface has shown that wood and wood-based products make up the majority of materials burnt, followed by polymeric materials. As such, the materials tested in the experimental burns included a number of wood-based and polymeric products commonly found in furnishings and home contents. Most of these materials were predominantly made up of carbon, oxygen and hydrogen. Because of this, their major emissions are CO2 and CO. However not all of the carbon is converted to CO2 and CO; the carbon that is not is released as aliphatic, aromatic and oxygenated hydrocarbons, some of which have the potential to harm people’s health.
Some of the materials such as particleboard, medium-density fibreboard, carpet and polyurethane foam contain a significant fraction of nitrogen. Burning these materials will produce hydrogen cyanide, ammonia, nitriles and nitrogenated hydrocarbons. All of these are potentially harmful to people’s health.
Emissions from all materials tested were evaluated against those from pine, a natural material widely used for structural construction. For this reason, it is a good reference material against which to compare manufactured and/or synthetic materials.
What Has Been Learnt?
Initial findings from the experimental burns, as well as the literature review conducted prior to the tests, show that exposures to compounds emitted during combustion of materials can lead to a range of adverse health impacts. These can be classified as follows:
1 Asphyxiants (CO and hydrogen cyanide).
2 Irritants (particles, volatile organic compounds, carbonyls, nitrogen dioxide, ammonia, hydrogen chloride and sulphur dioxide).
3 Impact on the central nervous system (CO, benzene, toluene, phenol).
4 Carcinogens (benzene, formaldehyde, naphthalene, isocyanates).
Some of these toxins, in particular chlorinated and nitrogenated compounds, are likely to be more present in bushfires at the rural/urban interface due to the types of fuels and materials burnt when compared with a forest bushfire. Other air toxins such as CO, particulate matter, formaldehyde and benzene are major air toxins emitted from both rural/urban interface and forest bushfires.
The results from the experimental burns are summarised in Figures 1 and 2, which show the emission factors (defined as the amount of a compound emitted per amount of fuel consumed) of CO, CO2, particulate matter, elemental carbon, organic carbon, and the elemental carbon to organic carbon ratio for 11 different types of combustible materials. Figure 3 shows the total carbonyls for these same materials. Results from a previous study that measured CO and CO2 emissions for CCA treated pine are included as a reference in Figure 1.
The highest emission factors of CO were observed for polystyrene, followed by plasterboard, particleboard with melamine and polyester (Figure 1). In comparison, pine recorded the lowest emission factors.
Fine Particulate Matter
The highest emission factors of fine particulate matter were measured for polyester and polystyrene (Figure 2). Both materials emitted approximately 20 times more particles compared with pine, while carpet emitted approximately nine times more particulate matter than pine. No major differences were observed among wood-based products.
Elemental Carbon & Organic Carbon
Wood-based materials were found to have the lowest carbon emissions, while polyester, carpet and polystyrene had the highest (Figure 2).
In general there was a larger fraction of elemental carbon compared with organic carbon. The exception was with polystyrene and plasterboard, where a significant organic carbon fraction was recorded. Lower elemental carbon to organic carbon ratios were also observed for manufactured wood products when compared with pine, with the difference likely due to the presence of glues and resins in the manufactured products. The organic fraction is likely to contain harmful organic compounds.
Carbonyls, a class of irritating volatile compounds, were emitted at highest concentrations during combustion of polystyrene, with emission factors for total carbonyls being 45 times higher than those for pine. Other materials with high emissions of carbonyls were plasterboard, polyester and polyurethane foams.
The relative distribution of individual carbonyls for each material is shown in Figure 3. It clearly shows, with the exception of polystyrene, that formaldehyde and acetaldehyde were the dominant carbonyls, contributing between 29 percent and 93 percent to the total emissions of carbonyls. Formaldehyde is a known human nasal carcinogen, while acetaldehyde is a possible human carcinogen. For the combustion modified polyurethane foam, acetaldehyde was the dominant carbonyl (80 percent). Both polystyrene and carpet had significant emissions of the strong irritant benzaldehyde. Acrolein, another strong irritant, was emitted primarily during combustion of wood-based products.
About two to four times the amount of emissions were observed for manufactured wood products compared with pine. A wide range of volatile organic compounds were identified during testing. These can have health effects that range from irritation of the eyes, skin, nose, throat and respiratory system, to headaches, dizziness, drowsiness and nausea, to being possible and known human carcinogens.
While a number of volatile organic compounds (benzene, toluene, styrene, xylenes, naphthalene, acetic acid and phenol) were present during combustion of all materials, others have only been identified during combustion of specific materials. These include pyrrole and nitriles, which were emitted during combustion of nitrogen containing materials only (particleboard, medium-density fibreboard, carpet) and pinene, camphene and limonene, which were emitted solely from combustion of wood-based materials. In general burning wood-based and polymeric materials released volatile organic compounds at higher concentrations than pine. As a result, the combustion of these products in fires at the rural/urban interface is likely to present a greater health risk than forest bushfires.
How Could the Research be Used?
The research will provide a better understanding of exposure risks to personnel and communities during bushfires that extend into the rural/urban interface.
The emphasis for assessing this exposure risk is on inhalation – exposures that occur outside of burning structures at varying distances from the fire and the smoke plume. It is assumed that in most cases firefighters do not wear breathing apparatus and that the majority of firefighting will be done outdoors. This clearly distinguishes firefighting at the rural/urban interface from firefighting at structural fires.
The outcomes from the research can be used to inform training, work practices and appropriate use of personal respiratory protective equipment, as well as to assess the use of truck-mounted air monitoring devices and to provide advice on firefighter and community safety.
Where to Now?
The emission factors determined during the small-scale experimental burns described in this article are not necessarily representative of all exposures firefighters could encounter. However, they do provide an important input into a high time-resolution dispersion model that will provide short-term modelled ground concentrations, resulting in a reliable estimate of potentially hazardous exposures.
Significant additional work is required to develop a useable set of scenarios and compare modelled exposure concentrations with previously measured exposures at structural fire incidents. Monitoring was conducted at structural training fires, which simulated room fires in three typical rooms (living room, bedroom and office). The results will provide data to validate outputs from the model.
Dr Fabienne Reisen is Bushfire CRC project leader and Research Scientist at CSIRO Marine & Atmospheric Research
For more information, go to www.bushfirecrc.com