OCR Text |
Show used to find the base carbon loss at a specified heat input per unit of furnace volume and coal volatile matter content. The base loss is multiplied by factors obtained from the two lower curves, accounting for effects of particle size and excess air. A curve showing the dependence of unburned carbon heat loss on excess air, derived from the B&W correlation for the coal volatile matter content, furnace loading, and coal particle size during the test under consideration, is shown in Figure 2. The observed losses, including those during the tests with the blended medium volatile bituminous coals, are approximately a factor of six higher than expected. Several possible contributions to the carbon losses shown in Figure 2 have been examined and compared with the measurements. A persistent problem has been difficulty in explaining how particles smaller than ..., 100 f.!m can survive passage through the furnace. Although it is possible that small char particles were generated during collection and size classification of the fly ash samples, it is assumed that the size distribution shown in Figure 3 is an accurate representation of the size distribution of char particles suspended in furnace exit gas. Of the models proposed, two were most successful in reproducing the observed particle size distribution. In one (Walsh et aI., 1993), char was divided into a reactive fraction, derived from vitrinite, and an unreactive fraction, derived from inertinite. The inertinitederived particles were assumed to have contained no volatile matter and to have a reactivity relative to the vitrinite-derived particles obtained from the measurements of Crelling et al. (1988). The other reasonably successful model assumed a low char reactivity, low furnace temperature (derived from the low furnace exit gas temperature mentioned above), and a highly stratified mixing-controlled oxygen concentration history in the postflame region (Walsh et aI., 1994). There are other mechanisms which may enable small particles to survive passage through the furnace and contribute to carbon loss, such as distribution of char reactivity (Hurt et aI., 1995), thermal deactivation (Davis et aI., 1995; Hurt and Davis, 1994; Hurt and Gibbins, 1995; Beeley et aI., 1995a), inhibition of char combustion by ash (Hottel and Stewart, 1940; Mitchell, 1991; Hurt and Davis, 1994), and passage of particles through regions of low temperature, near the wall of the furnace. The model for char combustion is derived from one developed by Walsh and Olen (1993) to describe unburned coke emissions from wall-fired boilers burning residual oil, later 11 |