OCR Text |
Show the atomizer is provided by a 50 cc syringe drive at a typical flowrate of 1 ml/min, although variations in flowrate of an order of magnitude can be obtained. Nozzle diameters of 0.5 and 0.7 mm have been used in the experiment. A plunger mechanism is connected to the atomizer (0.4 mm dia) so that if the fuel nozzle becomes plugged, the plunger can be inserted to clear the fuel feed orifice. The atomizer provides a conical spray which feeds into a converging section of the reactor leading to the reactor slot exit. In effect, this arrangement selects a slice of spray across the entire diameter of the spray cone, thus providing a representative sample of the entire spray distribution. At the same time, this method produces a uniform spray flow throughout the burner slot. The bulk of the spray contacts the walls of the prechamber and is discarded through a vacuum sump system. We are able to produce steady feedrates over a period of up to 30 minutes, at which time the slot exit must be cleared of soot buildup, and the atomizer requires cleaning. The photograph in Figure 2 shows a typical flame condition obtained in the reactor. With this design, we are able to produce a stable slot flame which matches the flow velocity of the main burner and thus we are able to define the temperature-time history of the slurry particles. Both temperature and oxygen concentration profiles along the reactor axis have been obtained and are shown in Figures 3 and 4. Temperatures were measured with an uncorrected 200 micron diameter Pt-Rh thermocouple. An estimated temperature correction for the thermocouple would be a maximum of 150K. The oxygen volumetric concentrations were determined by an Applied Electrochemistry Oxygen analyzer, S-3A. The axial temperature drops slightly, but is essentially constant up to 8 cm of reactor length, corresponding to a residence time of about 35 milliseconds. Similarly, the oxygen concentration is uniform over the first 8 cm and then increases at the point where turbulence entrains additional oxygen from the atmosphere. Thus we have a sufficiently uniform set of reaction conditions over the first 35 milliseconds of residence time. In addition to the nearly uniform axial profiles, the measurements show very uniform temperatures and oxygen concentrations both transverse to the slot and within the slot region (not shown) of the burner where size distribution measurements have been obtained. Equilibrium calculations of the temperature and oxygen concentrations for a methane-air flame have been performed and agree well with the measured values obtained by the oxygen analyzer and with corrected thermocouple temperatures. Given this range of operating conditions, the reactor system is able to simulate many of the conditions occurring at early times in large-scale combustion systems, including vaporization, devolatilization, ignition, and the early stages of char burnout. All the results discussed in later sections were obtained for the reaction conditions described above. OPTICAL DIAGNOSTICS The in situ PCSV system uses two independent laser beams and independent detection systems (Holve, 1983), and is capable of particle size measurements in the nominal size range of 0.3-120 microns, at high number densities (106/cm3). In addition, the instrument obtains the particle velocity distribution for all particles larger than approximately 10 microns. The median value of this size-integrated distribution is used to obtain the absolute particle number and mass |