OCR Text |
Show 14 be oxidized if the deposit remains in the combustion chamber any length of time. This is probably the reason that the deposit is found to have only 0.5 to 1/6 carbon after collection and cooling, while the postulated mechanism depends on the capture of carbon in the deposit to produce reducing conditions. Finally, only the surface layer of the deposit may be expected to remain liquid; because of the heat transfer to the underlying metal there will be a temperature gradient through the thickness of the deposit, and thus material below the surface will gradually solidify. The formation of the iron-rich crystals probably occurs during this phase transformation. As one region of the deposit cools, the iron can be reduced by the carbon monoxide present and withdrawn from the slag melt, where it crystallizes on the slag surface. This could occur in much the same way that elemental iron is formed from ore in a blast furnace or elemental iron is formed in the slag pool of a slagging combustor. The withdrawal of iron from the slag mass should cause solidification of the remaining slag. Unfortunately, the size and location of these crystals makes exact identification difficult. Future work on identification of the crystals, primarily through Mossbauer spectroscopy, which is capable of distinguishing between the various iron compounds in the sample, may shed more light on the exact mechanism of formation. Temperature Effects The importance of temperature on deposit microstructure can be seen in Figure 28. The fused nature of the sample in Figure 28a (along with the slag ,mass in Figure 26) is common in deposits collected at gas exit temperatures of 1200 C. However, in the case of Flame 2, additional heat extraction in the combustion chamber caused the gas exit temperature to fall to 900 C and resulted in a loosely packed deposit of lightly bound ash particles and cenospheres (Figure 28b). The peak temperature in the flame was apparently high enough to melt some of the ash into spheres and cenospheres as can be seen in the figure, but the lower gas temperature was sufficient to prevent a progressive liquid layer from forming and thus a smaller deposit was formed. The deposit was only about 20 to 25^ of the mass collected at higher temperature conditions, and was quite fragile. Figures 28c and 28d show the EDAX scans of the samples of 28a and 28b, respectively; these scans are almost identical, with the lower temperature sample having some calcium present. o , A deposit sample collected at a gas exit temperature of 1400 C is quite similar to that seen at 1200 c (See Figure 29). In this figure, the very smooth area across the bottom of the micrograph is the silver paint used to attach the sample to the holder for the SEM. The center area is the slag mass and the thin layer bordering the top of the slag is part of the Hastelloy X tube that was pulled away from the tube by the slag when it was removed. The slag appears to be very porous and to have been completely molten during formation. The close-up shown in Figure 29b also contains the tiny crystals of an iron-rich compound on the slag mass, located in the very center of the micrograph. Increased temperature also increased the severity of attack of the slag on the metal tube. In this study, s Hastelloy X tube was used; Hastelloy X has a composition of 22/6 chromium, 19/6 iron, small amounts of cobalt and molybdenum, with the balance being nickel. The deposit collect-i |