254
A. BHATT ET AL.
formation, is estimated to range from 0.002–0.72 lb/ MMBtu based on our surveyed data. Burning wood residues results in the highest CO emission factor, 0.72 lb/MMBtu, which is 20% higher than the value listed in AP-42. Burning a mixture of wood residues and sludge could result in a reduction of CO emis- sions by a factor of 1.5 as compared to AP-42, whereas burning biomass waste can further reduce CO emissions by a factor of 2.7 compared to AP-42. This is because increasing the air flow for oxidation of CO to CO 2 at high temperature is expected to result in high thermal NO x formation. Our analysis suggests that emission factors of organic compounds align reasonably well with the AP-42 emis- sion factors, with a range of 0.003–0.019 lb/MMBtu. Although the data sources reported higher emission factors for some HAPs, most of them are within the values specified in AP-42, providing a good agreement with the published literature. However, the median emission factor for HCl is much lower than the AP-42 reported data. It should be noted that the secondary data we use has a large variability, with a coefficient of varia- tion of 245%, which is likely due to the use of different particulate control technologies. Conclusion This analysis aims to compile and analyze emission factors from uncommon data sources for boilers that burn a single unconventional biomass-derived fuel or a combination of such fuels for which pub- lished data is scarce. The only emission factors that are currently available in the public domain and utilized by regulatory agencies are from EPA’s AP- 42 emission factors for wood residue combustion, which do not necessarily resemble the biomass- based fuels being utilized in industrial boilers. Therefore, using the AP-42 emission factors, could result in inaccurate estimates of potential emissions from industries that utilize boilers that burn uncon- ventional biomass fuels. Given that a chemical facil- ity that includes a biomass fuel boiler will need air permits with detailed emission calculations prior to its construction and operation, this study surveyed a broad knowledge base to identify data that could provide a better understanding of how the emis- sions in permits/permit documentation, stack test- ing, and relevant technical reports compare to those from AP-42. Upon a thorough review of emission factors from several data sources, which provide insights into emis- sion factors for biomass boilers, we analyzed the data
across the most important pollutants likely to be emitted from the boiler. The results suggest that the emission factors are highly variable for several air pollutants, depending on the type of fuel being combusted, the chemical composition of the fuel, and the fuel properties in the boiler. In general, the distribution of the emission factors from the studies reviewed in this work vary based on the type of fuel being combusted, with a right skewed distribution for filterable PM, PM 10 , PM 2.5 , SO 2 , and HCl. The median emission factors for NO x and the three PM components accounting for most of the emitted aerosol mass were about the same as the AP- 42 emission factors. For SO 2 , the median emission factor is 20 times higher than the AP-42 emission factor. For CO, the AP-42 emission factor is four times higher than median emission factor and is far outside the 95 th per- centile data obtained from the new sources. Our analysis also indicates that major toxic air pollutants (HAPs) have a higher AP-42-based emission factor compared to the median emission factor reported in this work. Use of these new emission factors based on fuel composition is likely to provide a better estimate from the complex feedstock combustion in the boilers when emission test- ing is not feasible. Although we provide a range of emission factors for a boiler that burns a mixture of biomass fuels, other process parameters should also be accounted for when selecting the best emission factors for each pollutant. These include the type of boiler utilized for combustion, flame temperature, and combustion practices employed at the facility. For example, fuel is combusted in stoker boilers by sitting on fixed grates, whereas fluidized bed boilers utilize inert par- ticles for air to blow through them so that the bed behaves as a fluid. Stoker boilers are typically utilized for large wood-fired units with high steam genera- tion, whereas fluidized bed boilers can handle fuels with more than 70% moisture content because of the larger share of thermal mass represented by hot inert bed particles. Similarly, the flame temperature and good combustion practices have significant effects on the products of combustion. This includes a proper air/fuel ratio, thorough mixing of air and fuel, and initial and sustained ignition of the mixture. Because air has a higher nitrogen content than oxy- gen, the required volume of air is much larger than the required volume of oxygen for a complete com- bustion. This affects the emissions of thermal NO x along with CO and CO 2 . Therefore, additional pro- cess parameters should be considered along with feed composition and fuel properties (such as moisture content) when selecting the most appropriate emis- sion factors for the boiler to be evaluated.
Made with FlippingBook Digital Publishing Software