Project at a Glance
Project Status: complete
Report Published April 1996:
Title: Atmospheric pollutant emission factors from open burning of agricultural and forest biomass by wind tunnel simulations
Principal Investigator / Author(s): Jenkins, B M
Contractor: UC Davis
Contract Number: A932-126
Topic Areas: Agriculture, Ecosystem Impacts, Impacts, Modeling, Stationary Sources
Atmospheric pollutant emission factors were determined by wind tunnel simulations of spreading and pile fires for eight different types of fuel including barley, rice and wheat straw, corn stover, almond and walnut tree prunings, and Douglas fir and Ponderosa pine slash. Cereal straws and stover were burned in fires spreading against an impressed wind induced by the wind tunnel blower. Pile burns in wood fuels were naturally ventilated through the side doors of the wind tunnel without assistance from the blower. Concentrations of gaseous and condensed phase species were monitored by sampling the constrained fire plume in the wind tunnel stack, 10 m above the tunnel floor. A conveyor system making up the floor of the wind tunnel was used with spreading fires to provide extended sampling times. Fires were allowed to spread naturally into a fuel bed moving continuously downstream on the conveyors. The flame position was held stationary in the tunnel by advancing the conveyor at a speed equal to the fire spreading rate. Loading rates were matched to published yields of crop residue. Wood fuels were burned in stationary piles of 35 to 45 kg each.
Measured were emissions of CO, NO, NOx, SO2, total hydrocarbons, CH4, non-methane hydrocarbons, total sulfur, CO2, particulate matter, volatile organic compounds (VOC), and polycyclic aromatic hydrocarbons (PAH). Filter and sorbent from one test with rice straw were submitted for analysis of chlorinated dioxins and furans. Size distributions of particulate matter were obtained via cascade impactor. Elemental compositions of particulate matter in the total, 2.5 pm and 10 pm size classes were analyzed to construct source profiles. Distributions of elements in particulate matter by size fraction were also evaluated using an eight-stage impactor. Bulk aerosol absorption coefficients were determined from Light transmission measurements through filter samples.
Results from spreading fires are presented for two configurations of the wind tunnel earlier determined to have greatest impact on particulate matter emission rates from rice straw. A configuration designated CEWF utilized an adjustable ceiling extended to the leading edge of the fire with an auxiliary floor inserted immediately below the fuel bed. The alternate configuration designated CRNF made use of neither.
Compared across the various fuel types, the emission factors by compound were often quite similar. Barley and wheat straw fuels produced higher levels of products of incomplete combustion, and spread more slowly as a result of reduced fuel loading rate compared to rice straw and corn stover. Enhanced levels of smoldering were associated with higher levels of CO and hydrocarbon emissions. In general, burning rates below five g s-l were associated with higher emissions of CO, total hydrocarbons, and particulate matter, and lower emissions of NOx and SO2. Although only small piles of wood fuels were burned, comparisons with full scale field data (where possible) suggest that wind tunnel results can be scaled appropriately by proper allocation of the mass of fuel burned in the flaming and smoldering stages. Particulate matter emissions are an exception. Wind tunnel results from smoldering with wood fuels were always less than published field values, but the number of wind tunnel experiments was limited and the comparatively small amount of material burned may have reduced the relative mass involved in smoldering compared to the field. Flaming stage results were quite similar to field results for fine particulate matter, but low for total particulate matter, most likely due to the lower total intensity and lower wind induced entrainment of the small, scale bums.
NOx emissions were heavily influenced by fuel nitrogen concentrations. Sulfur analyses showed a substantial amount of variation in the partitioning of sulfur among ash, SO2 in the gas phase and particulate matter. Barley straw was unique among cereal crop residues in producing large amounts of sulfur in particulate matter, principally as sulfate. Total sulfur analyses of stack gas were inconclusive as to the presence of other gaseous sulfur species due to instrument interference from CO2, Carbon, potassium, and chlorine dominated particulate matter compositions. Carbon was present principally in organic forms, potassium and chlorine in ionic forms. VOC emissions from Douglas fir and Ponderosa pine slash were high in comparison to the other fuels. PAH emission factors were higher for wheat straw and barley straw than for other fuels. Bulk aerosol absorption coefficients from filter samples varied from about 1 to 5 m2 g-l, with an average of 2.7 m2 g-l. Dioxin and furan analyses detected small levels of 1,2,3,4,6,7,8- HpCDD, 1,2,3,4,6,7,8,9-OCDD, and 1,2,3,4,6,7,8,9-OCDF but no TCDD was found above the detection limit.
Factors significantly affecting the spreading rates and to some extent the emission factors for cereal fuels were inlet air temperature, loading rate, and wind speed (as inlet air mass flow rate). Burning rate variations correlated well with air temperature and loading rate. Relative humidity was less significant in influencing fire spread. Inlet air temperature was observed to increase spreading rate in rice straw by 0.05m min-1 K-1. Increased wind speed was observed to slow fire spread in the CEWF configuration. Emission factors for CO, hydrocarbons, and particulate matter were higher and NOx, and SO2 lower in virtually every CEWF experiment when compared to the CRNF configuration with the same fuel. For particulate matter, this result was contrary to observations from earlier studies with rice straw, and is believed to be due to differences in the physical characteristics of the fuel, but requires further investigation. Averages of the two configurations were computed as a means of applying the results to the determination of emission offset allowances for biomass utilization facilities. More controlled experiments are recommended for understanding the influences of fuel and ambient conditions on fire behavior and pollutant formation.
For questions regarding research reports, contact: Heather Choi at (916) 322-3893
Stay involved, sign up with ARB's Research Email Listserver