| OCR Text |
Show Particulate Capture The removal of metal particles, associated with entrainment and aerosol dynamics, from the gas stream by flue gas cleaning equipment must be determined. The efficiency of particle removal typically depends on the particle size and on the design and operation of the flue gas cleaning equipment. Efficiency curve for a hots ide electrostatic precipitator (ESP) was utilized to predict the percent of metal capture as a function of particle size. The percent of metal capture depends on the fraction of metals condensed on each particle size; for example, volatile metals may condense largely on submicron particles generated by metal nucleation. Since most APCDs are least efficient in capturing submicron particles, volatile metals have a higher chance of escaping particulate removal devices. Combining calculations from the vaporization and entrainment step, aerosol dynamics process, and particulate capture, allows the estimation of controlled metals partitioning in the coal-fired combustion system. There are other modeling studies which have been recently conducted to predict trace elements behavior (particularly metals) from coal-fired boilers. 10kiniemi et al. [20] developed a numerical model (ABC Code) to solve simultaneously the dynamics of aerosol particles and volatilized metals and presented a comparison of simulated and measured results for aerosol and lead species behavior in PCC boilers. Their prediction of lead speciation due to reaction/ vaporization and aerosol dynamics corresponded closely to this study and other EER studies which predicted lead speciation and behavior in pec boilers. This is not surprising since 10kiniemi et al.'s modeling approach is based on the same principles used in this study's modeling approach, including the assumption of chemical equilibrium at combustion conditions and the use of fundamental aerosol dynamics equations to predict nucleation/condensation of metals. While their study stopped short of predicting particulate captUre at the APCD and metals emissions rates at the stack for pulverized coal-fired boilers, the similarity in methodology and in results adds credence to the both modeling efforts. DISCUSSION Figure 3 shows a comparison between predictions of the metals partitioning model and the experimental data ofMeij [21]. It is clear that the model does a reasonable job of predicting the distribution of species between the four fractions, even for metals that often are highly volatile, such as lead and cadmium. The model predictions for mercury are not surprising, since the metal is largely in the vapor phase and is not likely to be affected by the ESP. There is one question raised by this comparison, however, and that deals with selenium. While the model predicts that 8 |
