OCR Text |
Show remain in a vapor phase in the conditioning chamber unless adsorbed by quench water droplets or by particulate matter.6 The potential for adsorption by quench water droplets was evaluated by performing a simple mass balance. Theoretically the greatest amount of CDD/CDF will be removed at the lowest ESP inlet temperature. The maximum measured quench water flow rate at this temperature was 122 gpm (7.69 Us). Approximately one-third of this water evaporates to cool the flue gas while the remainder falls to the sump below the conditioning chamber. If the CDD/CDF is removed from the flue gas by absorption into the water droplets, then these compounds should be present in appropriate concentrations in the sump tank liquid. That water was not sampled in the current program but in an earlier test the CDD/CDF concentration was below the maximum detection limit of 68 ng/L. Based on the known quench water flow and by assuming that the CDD/CDF in the sump water is just below the detection limit, it can be calculated that the quench water will remove approximately 21 J.l.g/min of CDD/CDF. The measured CDD/CDF reduction rate across the conditioning chamber was 2277 J,lg/min. Therefore, absorption of CDD/CDF by the quench water accounts for less than 1% of the observed reduction. A similar result is obtained by assuming that the quench water is saturated with CDD/CDF. Based on these considerations absorption of CDD/CDF by the quench water was eliminated as an explanation for the observed reduction across the conditioning chamber. If the CDD/CDF is formed by gas-phase reactions in the combustor and it is absorbed by particulate in the mixing and conditioning chambers, the reduction in CDD/CDF across the conditioning chamber would be expected to be proportional to the removal of particulate surface area. Alternatively if CDD/CDF is predominantly formed on the surface of fine particulate then the CDD/CDF reduction across the conditioning chamber could also be expected to be proportional to the removal of particulate surface area (not mass removal). The PM loading at the furnace exit was not measured. However, both the PM mass loading and size distribution were determined at the ESP inlet. Based on an assumed PM loading from the furnace [approximately 3 gr/dscf (6900 ng/dscm)] the particulate mass reduction appears to be on the order of 60 to 80%. This is consistent with the fact that simple water sprays are relatively poor particulate co~trol devices. Extrapolation of particle size distribution data at the ESP inlet shows that approximately 95% of the remaining mass is smaller than 10 J,lm and that there are essentially no remaining particles larger than 20 J,lm. By assuming that the particles are spherical, that particles have constant density, and that all particles larger than 20 J,lm were 10 |