Quantification of Particle Number and Mass Emission Factors from Combustion of Queensland Trees ARINTO Y. P. WARDOYO, LIDIA MORAWSKA,* ZORAN D. RISTOVSKI, AND JACK MARSH International Laboratory for Air Quality and Health, Queensland University of Technology, GPO Box 2434, Brisbane, Queensland 4001, Australia The quantification of particle emission factors under controlled laboratory conditions for burning of the following five common tree species found in South East Queensland forests has been studied: Spotted Gum (Corymbia citriodora), Blue Gum (Eucalyptus tereticornis), Bloodwood (Eucalyptus intermedia), Iron Bark (Eucalyptus crebra), and Stringybark (Eucalyptus umbra). The results of the study show that the particle number emission factors and PM 2.5 mass emission factors depend on the type of tree and the burning rate. For fast burning conditions, the average particle number emission factors are in the range of 3.3- 5.7 × 10 15 particles/kg for woods and 0.5-6.9 × 10 15 particles/kg for leaves and branches, and the PM 2.5 emission factors are in the range of 140-210 mg/kg for woods and 450-4700 mg/kg for leaves and branches. For slow burning conditions, the average particle number emission factors are in the range of 2.8-44.8 × 10 13 particles/kg for woods and 0.5-9.3 × 10 13 particles/kg for leaves and branches, and the PM 2.5 emissions factors are in the range of 120-480 mg/kg for woods and 3300-4900 mg/kg for leaves and branches. Introduction Biomass burning including controlled and uncontrolled forest and savannah fires, as well as various types of residential burning, has been identified as a major contributor to particles and gases in the atmosphere (1-3). These particles and gases impact human health and are linked to morbidity and mortality (4), and play a significant role in affecting atmospheric processes (5) such as radiation balance (6) or acidification of clouds, rain, and fog (7). In particular, forest fires significantly contribute to the particle burden of the atmosphere. More than 1.3 million hectares of forest were burnt in China in 1987. In the same year, forest fires in eastern Asia consumed approximately 14 million hectares (2). In 1994 and 1997, forest burning destroyed more than 50,000 and 20,000 km 2 , respectively, in Indonesia (7). Other data show that 100,000 km 2 of forest in northern latitudes, 400,000 km 2 of tropical and subtropical forest, and 5-10 million km 2 of open forest and Savannahs are burnt every year (3). In Australia, the state of Queensland recorded its the worst forest fires in 1991 with 37,000 hectares consumed (8). From July 2002 until June 2003, there were 2,618 fires in this states covering one million hectares in this state (9). In 2004, the major fires occurring within the southeast corner of Queensland included forest fires at San Fernando and Canugra in July; at Wallaby Hill Mudgeeraba, Gold Coast, Minden, Gilston/Tallai Range, and Tamborine in August; and at Tamborine, Lowry/Hinze Dam, and Nerang in October. Knowledge of particle emission characteristics in terms of size distribution and emission factors, particularly in terms of particle number emissions, has been identified as a very important element in developing quantitative assessment of the impact of the fires. Both these characteristics depend on the type of biomass and the conditions of burning. Studies reported in the literature showed that the majority of particles resulting from biomass burning were less than 2.5 μm in diameter (10-13). The PM2.5 emission factors have been measured in the range of 0.2-12 g/kg (12, 14, 15). In terms of particle number, the reported emission factors ranged from 3 × 10 15 to 40 × 10 16 particles/cm 3 (13). However, the existing data on particle size distribution and number emission factors are still very limited, with data unavailable for many tree species, such as those growing in the frequently fire-ridden state of Queensland. This paper presents the results of a controlled laboratory study aimed at quantifying the particle emission factors from combustion of trees typically found growing in southeast Queensland open forests. A specific emphasis of the study was on developing a better understanding of the size distribution and emission factors from burning conducted under different environmental conditions. Experimental Section Quantification of the emission factors and measurements of particle size distribution were conducted by the burning of biomass samples in a stove modified for the purpose of the study. The sampled smoke was diluted in two steps, first with compressed fresh air in an ejector dilutor (Dekati) and then by mixing with filtered air. Particle number emission factors were quantified by measuring the total particle concentration during the combustion process using a condensation particle counter (CPC), while the size distribu- tion of particles was measured using a scanning mobility particle sizer (SMPS). PM2.5 emission factors were ap- proximated using a TSI Dustrak with a 2.5 μm inlet. Experimental Setup. The experiment was designed and carefully controlled to capture the maximum number of parameters, which can be controlled under laboratory conditions. In particular, the flow rate for the experiments was chosen to correspond as closely as possible with the flow rates under natural conditions, represented by wind speed. It has been reported in the literature that most fires occur at a wind speed between 70 and 120 km/h. Bush- fires in Australia occur under typical wind speed of about 80 km/h and with a rate of spread of 18-20 km/h (http:// www.ffp.csiro.au). The experiments were set up to simulate burning conditions by injecting air with the speed of 20 m/s (72 km/h) into the stove for fast burning and by keeping the stove unconnected to the blower during slow burning, so as not to force the air supply through the ventilation system (unforced flow rate of the incoming air was in the range between 1.7 and 2.5 m/s). The performance of the measurement system was in- vestigated by adjusting the flow rate of air in the dilution tunnel and the temperature of the heated air in the Dekati diluter. In particular, experiments were conducted for several constant values of temperature of the injected air and for different flow rates of air in the dilution tunnel. The results * Corresponding author phone: + 61 7 3864 2616; fax: + 61 7 3864 9079; e-mail: l.morawska@qut.edu.au. Environ. Sci. Technol. 2006, 40, 5696-5703 5696 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 18, 2006 10.1021/es0609497 CCC: $33.50  2006 American Chemical Society Published on Web 08/16/2006