OCR Text |
Show A useful strategy for reduction of NOx in such burners is the use of inserts, which probably act by simply cooling the flame. This may, however, be accompanied by an increase in other emissions such as CO and CH20. We plan to enlarge our LIP Bunsen flame studies described here to include flames with various inserts. A burner has been designed to provide a stable, partially premixed flame. This is important for several reasons. A temporally stable flame is more reliably modeled than a flickering flame, and data may be acquired averaging over many laser pulses, thereby improving signal to noise ratios. The burner consists of a steel tube providing the inner flow, with an i.d. of 1.65 cm and o.d. of 1.9 cm. This is surrounded by a coaxial air flow of 20 cm diameter. This large diameter is needed for stability in the upper region of the flame, which can be about 15 cm high. Two layers of honeycomb straighten the flow. For the given conditions, flames could be stabilized for premixed stoichiometries of <1> -0.7 and also <1> > 1.2. The inner cone height depends strongly on <1>. For <l> between 1.2 and 1.5, there is no visible flickering of the inner cone. OH and CH distributions are determined in the Bunsen flame using LIF. OH is an important radical marking the different reaction zones, so its distribution provides important information about the flame structure. Excitation scans in OH determine temperature. The measurement of CH locates the regions of formation of prompt NO, near the premixed flame front LIF Measurement of OR Radicals For the OH measurements, the laser beam near 308 run wavelength was focused to a spot size of 100 J.Ull x 200 J.Ull. Spatial distributions were determined using a particular rotational line, exciting a level whose fractional population (Eq. (3» is relatively insensitive to temperature. Temperatures were measured by excitation scans like those in Fig. 3, using 17 different lines. Although quenching and hence fluorescence quantum yield, Eq. (1), may vary in different regions of the flame, we estimate differences 6 of only -20% through the flame and have ignored such differences in the analysis. This may not be valid in the outer region of the diffusion flame which is colder and thus denser, but little OH exists there. Relative OH concentrations were measured for different stoichiometries of the premixed flame. Vertical proftles were determined along the axis of the flame. The profiles needed to be corrected for the absorption of the laser by the OH as the beam passed through the flame. Also, some absorption of fluorescence light occurred, and also was accounted for. These absorption phenomena must be considered here because the total number density of OH is high at atmospheric pressure; it was not an issue in the low pressure flame. 13 |