252
A. BHATT ET AL.
Summary of emission factors Box plots in Figure 5 show the emission factors from various data sources, including air permits, stack test data, and NCASI technical bulletins, compiled and ana- lyzed in this research. For the aerosol species, the dis- tribution of uncontrolled emission factors is skewed to the right, and the mean emission factor is an order of magnitude greater than the median value. Also shown in Figure 5 are the emission factors currently included in EPA’s AP-42 Chapter 1.6 emission factors (marked as red dots). Interestingly, the emission factors from EPA’s AP-42 are slightly less than the median for filterable PM, but almost same as the median calculated from the data collected for this work for both PM 10 and PM 2.5 . Because condensable PM is smaller than 2.5 microns in diameter, the emissions of the condensable component of PM are very small compared to PM 10 and PM 2.5 . For the gaseous pollutants, the emission factor for NO x from EPA’s AP-42 is 12% lower than the calculated median. However, the emission factor for CO in AP-42 is higher by a factor of four compared to the calculated median emission factor. The emission factor for SO 2 in AP-42 is smaller by a factor 20 compared to the calcu- lated median emission factor, which is due to the high sulfur content in biomass fuels as compared to wood residues. Among the HAPs, the median emission factor
for acetaldehyde aligns well with EPA’s AP-42 database, but the calculated median emission factor for formalde- hyde is about a tenth of the emission factor used in EPA’s AP-42 database. The median emission factors for two acid gases, HCl and HF, are only about 59% and 67% of those in AP-42, respectively. There is a large range in the emission factors for HCl, and that is likely due to the type of particulate control technology. Wet scrubbers are highly efficient at capturing HCl com- pared to dry particulate control technologies. Table 4 compares the emission factors for several pollutants from AP-42 to the range of emission factors and median numbers from the new data sources, disaggregated by type of fuels. Note that the plots in Figure 5 are based on uncontrolled emission factors and thus do not include CORRIM data (control technology information was not available from CORRIM to allow us to calculate uncon- trolled emission factors). The results show that a boiler burning wood and wood residues with high moisture content likely emits more PM than a boiler burning dry fuel, whereas a boiler burning 100% sludge has lower PM emissions than a boiler burning wood residues. For combustion of biomass-derived waste fuels, we find PM emission factors to be higher than wood residues and sludge. Our results show that the
Figure 5. Distribution of biomass-fueled boiler emission factors from collected data sources for each pollutant and inclusive of all fuels. Whiskers correspond to 5 th and 95 th percentiles, with means shown as black squares, medians as horizontal lines, and emission factors from EPA’s AP-42 as red dots. The number of data points used for each pollutant are also shown (n). These box plots report uncontrolled emission factors and thus do not include the CORRIM data.
Made with FlippingBook Digital Publishing Software