OCR Text |
Show 1.3.7 pyrex spiral condenser and a series of three water traps were used to remove moisture. Further, particulates were trapped with a series of filters. The instruments were calibrated with special gas mixtures. All the instruments were checked for possible interference errors and were found to be free of them. As the continuous sampling method was used and the measurements were performed allowing enough time to achieve consistency of readings, the errors because of adsorption of species in the sampling train were expected to be negligible. As the outer diameter of the probe was small (less than 4% of the width of the flame) and quartz was the material, the errors caused by aerodynamic disturbances and heat abstraction attributable to the probe were negligible. The probe was blown clear of soot frequently between runs and the filters were replaced. Hence, the errors because of the catalytic conversion of N0? to NO in the sampling line were minimized [14]. As quartz probes have been used successfully by other investigators for SO- sampling [15], SOp concentration measurements were considered satisfactory. As the gases cooled rapidly in the sampling line, the oxidation of hydrocarbons and soot inside it did not affect the concentration of CO. For the same reasons, the oxygen concentrations were not expected to change in the sampling line. Flame temperatures were measured with a silica coated Pt/Pt-13/o Rh thermocouple (bead dia = 0.25 mm). The measured temperatures were corrected for conduction and radiation losses following the equations given by Fristrom and Westenberg [16]. The corrections ranged from 50 to 130 K. The fuels were ASTM No. 2 oil, and SRC-II (a 5.75:1 mixture of middle and heavy distillates). The emulsions were prepared by initially mixing the oils and water with a high-speed blender. As the emulsions |