OCR Text |
Show ,.--- .' .. .. .... .- decreasing (Figure 4). The probable reason for this behaviour will be discussed next. 5. Discussion 5.1 Variability The non-similar nature of the temperature-time traces seen in Figures 1-4 is, at first, quite unexpected since all the particles in those experiments were drawn from the same sample. Not only do they burn with different burn times and temperatures, the shapes of their temperature profiles are different. This has important consequences. Often, in modelling, we rely on 'averages'. Clearly in the present situation, 'averaging' among particles is not possible and even if one were to use some kind of average burn time or temperature, the physical meaning of such an average is not obvious. It has been shown [7] that the different traces can be satisfactorily explained if differences in the initial particle size ( and shape) and chemical rate coefficient are properly accounted for. Indeed, given the heterogenous nature of the char, it is quite plausible, that, not all of them should burn at the same rate. There are differences in the external shape, effective size, internal structure, and mineral matter distribution and composition between particles drawn from the same sieve cut. These differences cause the observed variations in the traces. To show that the observed variations are not due to experimental error, similar experiments were done with synthetic chars made in the laboratory [8]. Since these chars are very homogenous and spherical, almost all of them burn with similar profiles. 5.2 Tumbling Figure 5 shows another curious observation from the experimental results. It is seen that there are regular oscillations in the intensity traces as the particle burns. The amplitude of the oscillations is larger than any noise in the system. It is clear that they cannot be due to some random phenomena since they are quite periodic in nature. The only explanation is that of particle tumbling. These traces are created 9 |