Soot and OH distribution at the smoke point of diluted propylene-hydrogen mixture diffusion flames in cross-flow

An experimental study of smoke point diffusion flames in cross-flow is presented from the perspective of understanding the role of hydroxyl (OH) radicals and oxygen molecules on soot concentration. The laser induced incandescence (LII) and the laser induced fluorescence (LIF) techniques were employed to determine soot and OH concentrations. A propylene gas jet flame from a circular tube burner (diameter=3.2 mm) with a range of exit velocities (4.2 to 34 m/s) corresponding to a Reynolds number range of 520 to 6065 was subjected to a cross-flow with velocities ranging from 2 to 4 m/s. About 10% (by mass) of hydrogen was mixed with propylene to keep the flame attached to the burner at all cross-flow, jet velocity, and dilution conditions. Nitrogen was added to the fuel stream to reduce smoke formation. This paper presents the measurements in flames with the critical mass flow rate (CFMFR) of pure fuel at the smoke point, and with different fractions of the CFMFR diluted with nitrogen to achieve the smoke point, in the cross-flow with a velocity of 3 m/s. The soot and OH concentration contours in the plane normal to the cross flow stream exhibit a kidney shape. In most instances, high OH concentration was measured at the outer rim of the flame, whereas soot concentration was noted to be highest at the center of the flame or middle of the kidney-shaped vortex. Overall, the flame at the 100% CFMFR had a higher soot concentration than the 60% CFMFR flame, except at 25% of the flame length away from the burner. On the other hand, the 10% CFMFR flame had a higher soot concentration than the 20% CFMFR flame. The overall OH concentration at 25% of the flame length of the 100% CFMFR flame was the lowest, and that in the 10% CFMFR flame was the highest. The flame trajectory and the momentum flux ratio, which govern the fuel-air mixing intensity, played an important role on the behavior of soot and OH concentrations. The transition from momentum to chemical domination of smoke emission when the fuel flow rate was changed from higher to lower CFMFR was governed by the change in the dependence of soot oxidation on oxygen to that on hydroxyl radicals. The hypothesis based on the distinction between the dominance of physico-chemical processes in chemical and momentum controlled regions found in flames in quiescent surroundings is valid also in flames in cross-flow.

S o o t a n d O H D i s t r i b u t i o n a t t h e S m o k e P o i n t o f D ilu te d P r o p y l e n e - H y d r o g e n M i x t u r e D if f u s io n F la m e s in C r o s s - F lo w S. F. Goh and S. R. Gollahalli Combustion and Flame Dynamics Laboratory School of Aerospace and M echanical Engineering University of Oklahoma, Norman, OK 73019, USA e-mail: [e-mail] Abstract An experimental study of smoke point diffusion flames in cross-flow is presented from the perspective of understanding the role of hydroxyl (OH) radicals and oxygen molecules on soot concentration. The laser induced incandescence (LII) and the laser induced fluorescence (LIF) techniques were employed to determine soot and OH concentrations. A propylene gas jet flame from a circular tube burner (diameter=3.2 mm) with a range of exit velocities (4.2 to 34 m/s) corresponding to a Reynolds number range of 520 to 6065 was subjected to a cross-flow with velocities ranging from 2 to 4 m/s. About 10% (by mass) of hydrogen was mixed with propylene to keep the flame attached to the burner at all cross-flow, jet velocity, and dilution conditions. Nitrogen was added to the fuel stream to reduce smoke formation. This paper presents the measurements in flames with the critical mass flow rate (CFMFR) of pure fuel at the smoke point, and with different fractions of the CFMFR diluted with nitrogen to achieve the smoke point, in the cross-flow with a velocity of 3 m/s. The soot and OH concentration contours in the plane normal to the cross flow stream exhibit a kidney shape. In most instances, high OH concentration was measured at the outer rim of the flame, whereas soot concentration was noted to be highest at the center of the flame or middle of the kidney-shaped vortex. Overall, the flame at the 100% CFMFR had a higher soot concentration than the 60% CFMFR flame, except at 25% of the flame length away from the burner. On the other hand, the 10% CFMFR flame had a higher soot concentration than the 20% CFMFR flame. The overall OH concentration at 25% of the flame length of the 100% CFMFR flame was the lowest, and that in the 10% CFMFR flame was the highest. The flame trajectory and the momentum flux ratio, which govern the fuel-air mixing intensity, played an important role on the behavior of soot and OH concentrations. The transition from momentum to chemical domination of smoke emission when the fuel flow rate was changed from higher to lower CFMFR was governed by the change in the dependence of soot oxidation on oxygen to that on hydroxyl radicals. The hypothesis based on the distinction between the dominance of physico-chemical processes in chemical and momentum controlled regions found in flames in quiescent surroundings is valid also in flames in cross-flow. Introduction Currently, the methods to reduce smoke emission from turbulent hydrocarbon jet flames depend upon the reduction of the soot formation rate or increase of its oxidation rate by the addition of diluents or entrainment of additional air into the flared gas stream. This study deals with mixing inert gases into the fuel stream to suppress the smoke. Goh et al. (2001) found three regions (momentum controlled region, chemicalcontrolled region, and a transition region bridging the two) on the curve depicting the variation of the amount of diluent gas required to suppress smoke emission versus fuel mass flow rate. The pure fuel mass flow rate at the smoke point was defined as the Critical Fuel Mass Flow Rate (CFMFR). A previous study by authors (Goh and Gollahalli, 2004) delineated the dominant oxidation mechanisms in the momentum and chemical controlled regions in hydrocarbon jet flames in quiescent surroundings. The main objective of the present study was to investigate the coupling effects, if any, of the cross-flow on the distinction between the momentum and chemically dominated regions. Laser induced incandescence (LII) method is an effective non-intrusive method currently used to measure soot concentration in flames. It had been applied in many areas besides diffusion flames, for instance internal combustion engines (Hofeldt, 1993), gas turbines (Brown and Meyer, 2002, Jenkins et. al, 2002) and shock tubes studies (Woiki et. al, 2000). It is superior to other non-intrusive methods like the extinction technique and intrusive methods like sampling followed by transmission electron microscopy (TEM). The LII technique applies the mechanism of heating of soot particles to a temperature higher than the surrounding gas temperature by the absorption of laser energy, and the subsequent detection of the blackbody radiation (incandescence), which corresponds to the elevated soot particle temperature. The spectrum of the emitted incandescence is broad and is a strong function of the soot particle temperature. The LII signal is linearly related to the local soot concentration for a certain range of laser fluence, hence only a point calibration is sufficient. Measurement on a short life species like OH can be performed with a non-intrusive method like laser induced fluorescence (LIF). The LIF finds the population density of selected quantum states and the thermodynamic properties of the test medium. A specific narrow band laser is used to selectively excite the atom or molecule to explicit vibrational and electronic levels from its initial state. The fluorescence emission from the excited species is proportional to the total density of the population of probed species at the ground state. Experimental Apparatus The fuel and the inert gas used in this study were propylene and nitrogen. A sharp-edged stainless steel 3.2 mm ID tube was used as the burner. All experiments were conducted in a vertical wind tunnel with a test section of length 76 cm and the cross-section of 35.6 cm X 35.6 cm. Two side walls were fitted with high optical quality plate quartz glass to allow ultraviolet radiation to pass in (laser beam) and out (flourescence), and the other two side walls were fitted with metal plates, of which one of them was used to introduce the burner. The smoke point determination method was the same as employed by Goh et al. (2001). A 5 mW Helium-Neon laser with a cylindrical lens was used to generate a light sheet (wave length = 633 nm) of 3 mm thickness about 5 cm vertically above the tip of the flame as a means to determine its smoke point conditions. When a flame started to smoke, the laser sheet became visible due to the scattered radiation, and the intensity of the illumination of the laser beam was a function of the concentration of smoke at that location. A viewing angle of about 15 degrees with respect to the laser beam in the forward scatter mode was found to yield repeatable and sensitive results. The smoke point of the flame was determined when the laser sheet just turned visible as a continuous sheet. A mixture of propylene and hydrogen was used as the fuel to keep the flame attached to the burner at all times and at all conditions of cross-flow and dilution. The direct premixing of hydrogen and fuel was preferred to a pilot hydrogen flame to avoid the complexities due to the bulkiness of annular pilot flames normally used in experiments in quiescent surroundings, which affect the fluid dynamics near the burner, particularly due to their large wakes in cross-flow. Since the near-burner fluid dynamics and the oxygen infusion have strong effects on soot formation, the pilot flames were undesirable for this study. In addition to the pure fuel experiments, the fuel mass flow rate was set at ten different values, 90%, 80%, 70%, 60%, 50%, 40%, 30%, 20%, 10% and 5% of the CFMFR and measurements were obtained.. At each of these fractions, the amount of nitrogen addition needed to render the flame smokeless was determined. LII Instrumentation The local soot concentration was measured using the LII method. The laser fluence for the LII measurement was 0.34 J/cm2. Since the LII and the LIF methods involved somewhat similar instrumentation, the schematic diagram of the experimental setup of both instrumentations is shown in Figure 1. LII method applied a Nd:YAG laser with a series of optical devices. The Gaussian profiled 1064 nm laser with pulse width of 8 ns was generated by a Nd:YAG laser operating under Q-switch condition. The detection setup was arranged perpendicular to the laser beam path. A high power-focusing lens was used to focus the beam to a size close to 1 mm diameter at full width half maximum (FWHM) value. A band pass filter with a central wavelength of 431 nm (bandwidth of 9 nm at half the maximum transmission) was used to reduce background noise. An integrated charge-coupled device (ICCD) camera with an array size of 576 x 384 pixels synchronized with the laser pulse was used to capture the image. A 30 ns gate was selected because it captured most of the rise period during the laser heating, and it also avoided the influence of size-dependent heat loss path because the heat transfer through vaporization was more important for population measurement in this period of time (Mewes and Seitzman, 1997). Image data were streamed to a PC based image acquisition system The LII signal was calibrated by duplicating the experiments of Shaddix and Smyth (1996). The laser was fired from the y direction (perpendicular to the jet axis) and the signal was measured in the x direction (parallel to the jet axis). The measurements were taken by traversing the laser beam in the x direction with one mm step size for the 25% CFMFR flame, and two mm step size for the rest of the flame conditions. H alf of the x-y cross section of the flame was covered at three flame locations at different distances from the burner exit. The results were averaged over 100 measurements. LIF Instrumentation The laser induced fluorescence (LIF) experimental setup was somewhat similar to the LII setup. The laser equipment used for LIF measurements consisted of another pulsed Nd:YAG laser coupled to an Optical Parametric Oscillator (OPO) with a Frequency Doubler Option (FDO). The same ICCD camera system with a different bandpass filter was used for detection of OH. A frequency-doubled output (285.26nm) of the FDO was used to induce fluorescence (Battles et al, 1994), which was detected at a wavelength of 315 nm. At this wavelength, OH was pumped at the Qi(6) transition in the OH A2'L^X ln system of the (1,0) band and the fluorescence from the (1,1) band was collected (Hanson, 1986). For both LII and LIF measurements, an excitation beam of diameter 1 mm was used. For a diffusion flame in atmospheric condition, only 3 out of 1000 excited OH radicals loose energy through fluorescence, and the rest of the molecules loose their energy through collisional quenching with ambient gases (Tamura et al. 1988). For the quantification of OH radical concentration, a commonly used method involves the direct estimation of the quenching rate at the location of the measurement to determine the quantum yield. This method requires the concentration field of species involved and is relatively simple and more feasible. The concentration and the local flame temperature can be obtained from the numerical results. The details of spectroscopic modeling given by Laufer (1996) with collisional quenching parameters of Tamura et al. (1998) were adopted in this study. The collision species used for the calculation were N2, O2, CO, CO2, H2, H2O, CH4, H and OH whose concentrations were obtained from the computational results using an equilibrium model. Since the numerical model predictions agreed well with the measured results for the major species (O2 and CO2) concentrations and flame temperature, they were considered reasonable for quenching rate calculations. Figure 1 shows the schematic of the combustion chamber and instrumentation. The nominal experimental conditions and uncertainties in the measurements are presented in Tables 1 and 2. Results and Discussion Figure 2 shows the variation of CFMFR and the nitrogen mass flow rate to achieve smoke point for all tested cross-flow speeds. A transition region (corresponding to the maximum nitrogen mass flow rate needed to suppress the smoke) separated two distinct regions, which were defined as "momentum dominated region" (right of the transition) and "chemical dominated region" (left of the transition). This behavior is similar to that found in experiments with quiescent surroundings, Goh et al., (2001). To get a better understanding of the thermochemical processes in these three regions, four flame conditions (CFMFR values) were selected for the detailed study. From Figure 2, we see that all flames at different cross-flow velocities have the transition region around 20% CFMFR. Hence, in this paper, the jet flow rates corresponding to 10% (chemical-dominated region), 20% (transition region), 60% and 100% (momentum-dominated region) of the CFMFR were selected for OH and soot concentration measurements. Figure 3 shows the plot of momentum flux ratio, R, which is PV 2 defined as R = ---, where p is the density, is the v velocity, and subscripts j and « refer to jet exit and crossP v A^co flow conditions, co at the smoke point with CFMFR. For a given cross-flow, as the CFMFR increases the momentum flux ratio increases due to the fact more air has to be entrained into the flame to render it smokeless. For a given CFMFR, the momentum flux ratio at the smoke point decreases when the cross-flow velocity is increased. More interestingly, the increase of cross-flow velocity shows a decrease of the gradient of these curves. The gradient becomes significantly small at cross-flow velocities above 3.0 m/s, implying the cross­ flow velocity above a certain value (in this case 3.0 m/s) has very little influence on the structure of smokepoint flames. These results are corroborated by Goh (2003) who showed that with the increase of cross-flow velocity form 2.0 m/s to 3.0 m/s, the cross-flow sheared the jet more intensely, penetrated into the jet, and thus impacted soot oxidation rate severely. However, above 3.0 m/s cross-flow, the flame bent over quickly in the near-burner region itself, changed the flame shape from that of a flame in a cross-flow to a flame in co-flow, and reduced the effectiveness of cross-flow on the flame. Hence, in this paper, the flames in the 3 m/s cross-flow were selected for detailed diagnostics. The hydroxyl radical (OH) and soot concentration were determined at the x-y plane cross sections of the flame at different distances from the burner exit. Figure 4 shows the locations of these planes. The concentration measurements are shown as tomographic contour plots in Figs. 5-9. Most x-y plane tomographic contour plots of the OH and soot concentration show a kidney shape, which is in agreement with the previous study on major product concentrations in bent-over flames (Gollahalli et al., 1975). This kidney shape was caused by the counter rotating vortex pair (CVP). The CVP initiated very near the burner exit by the interaction between the jet and the cross-flow. The shear layer folds and rolls up the jet and this phenomenon was documented in several studies (Kelso et al., 1996). The CVP expands due to the entrainment of surrounding fluid and dominates the downstream flow. Figures 6 and 8 show a high soot concentration at the center of each vortex ring. The higher air entrainment was documented at the center of the flame (y=0) by Goh (2003), where air entered in the positive x direction through the process of rolling and folding of vortices. Because of the low concentration of oxidizer (both OH and O2 ) in these vortex rings, the soot particles form and grow in that region. Figure 5 shows the OH concentration at 25% of the flame length for all four flame conditions studied. The OH concentration was high in the vicinity of the outer rim of the flame and in the vicinity of y=0 region. This high OH concentration corresponds to the locations where chemical reaction rate is high. The high OH concentration near the flame centerline (y=0) was mainly due to the CVP effect, which enhanced the fuel-air mixing and reaction rates. The OH concentration was not measurable at downstream locations. The high soot concentration in the flame caused a high emission of broadband radiation, and resulted in a low signal to noise ratio. Also, the OH concentration at downstream locations was too low for detection as corroborated by the equilibrium computations. The second reason was that OH was an active participant in the soot oxidation process. The scarcity of OH may also be due to a high rate of soot oxidation by OH radicals. There was ample evidence in the study of flames in quiescent surroundings (Goh and Gollahalli, 2004) supporting the hypothesis that the soot in the chemical dominated region was mainly oxidized by the OH radicals. On the other hand, in the momentum dominated region flame, the soot oxidation was dominated by the oxygen concentration.. The two main players in the soot oxidation process for a diffusion flame are OH and O2 (Fenimore and Jones, 1967). From the perspective of soot oxidation, since the supply of OH is chemical rate dependent and that of O 2 is diffusion rate/momentum flux dependent, it is reasonable to identify the chemical and momentum dominant regions at different CFMFR values in cross-flow. Furthermore, at the 25% flame length for all four flame conditions, we notice that the OH and the soot concentration distributions show opposite trends. The results in the 10% CFMFR flame (Figure 6) show that the highest soot concentration points were located at around 75% of the flame length; The change of flame structure due to the decrease of the momentum flux ratio caused by lowering of the fuel jet flow rate at the same cross-flow velocity, increases the flame residence time, which was substantiated by the flame length measurements (Goh, 2003). The increase of the residence time allows the soot generation over a longer period of time and shifts the peak soot concentration point downstream. The increase of momentum flux ratio due to the higher fuel jet velocity at the same cross­ flow velocity, on the other hand, causes the high soot concentration point to shift upstream. Figures 7 and 8 show that in the 20% and 60% CFMFR flames the peak soot concentration points are located at 50% of the flame length instead of 75% in the 10% CFMFR flame. In the 100% CFMFR flame, the soot concentration magnitude at 50% and 75% flame length were close. The higher momentum flux ratio causes a higher flame gas mixing, which narrows the difference of the soot concentration at these two locations. In the momentum-controlled flames (60% and 100% CFMFR), the soot concentration contours at 25%, 50% and 75% flame lengths, show a kidney-shaped structure (Figures 8 and 9). Besides the reason of the oxidation of the soot particles by OH, which was discussed previously, the high soot concentration in the core of the vortices could also be due to circulating flow in the vortices that bring hot combustion products into the core while flinging relatively colder air into outer regions. Hence, the insufficiency of oxidant and longer residence time (due to the circulation inside the flame) contribute to the higher rate of soot production, and cause a higher soot concentration in the cores of vortices. In most regions, the soot concentration in the 100% CFMFR flame is comparable to that soot in the 60% CFMFR flame (Figure 8 and 9), except at 25% flame length where it is slightly lower in the 100% CFMFR flame. At 25% of flame length, the soot concentration in the 20% CFMFR and 10% CFMFR flames was very close (Figure 6 and 7). The difference in soot concentration between the 100% and 60% CFMFR flames was not as significant as that between the10% and 20% CFMFR flames, because the former set of flames were at a higher momentum flux ratio than the latter. A higher momentum flux ratio brought in larger amount of O2 into the flame, and consequently reduced the overall level of soot concentration and the difference in soot concentration between the momentum-dominated flames. On the other hand, at a lower momentum flux ratio, the overall soot concentration was higher. The flame structure of 10% CFMFR flame was somewhat similar to that of a laminar flame in co-flow (Goh, 2003). The cross-flow simply bends the flame and aligns it with the air stream, and thus turns it into a flame in a co-flow. Thus, the cross-flow has much less effect on this flame than on the momentum-dominated flame. Due to the lower mixing with the air, the flame has to rely on the OH radicals to oxidize the soot. The previous study on flames in quiescent surroundings by Goh and Gollahalli (2003) showed that the oxidation rate of soot by OH was higher than by the O2 molecules for most part of the chemical-dominated flames. However, OH-soot oxidation rate was only significant at the region very close to the burner in the momentum-dominated flames, and O2-soot oxidation dominated most part of the flame. This paper demonstrates that behavior is retained even in cross-flow. Conclusions The variation of diluent flow rate required to suppress smoke emission with fuel flow rate in jet flames in cross-flow also shows three distinct regions (chemical and momentum-dominated regions, and transition regions) similar to the trends previously documented in flames in quiescent surroundings. At a higher jet/cross­ flow momentum flux ratios, the cross-flow has a smaller effect on smoke-point conditions of the flame. The hydroxyl radical and soot concentration measurements in the flames at different fuel flow rates in the aforementioned regions show significant differences between the chemical and momentum dominated regions. In a chemically dominated flame, hydroxyl radical dominated the soot oxidation rates and smoke point due to the low mixing rate. However, oxygen dominated the soot oxidation in the momentum dominated flame. Since OH is chemical reaction rate dependent and oxygen is diffusion rate dependent, the hypothesis on the distinction between the dominance of physico-chemical processes in chemical and momentum controlled regions found in flames in quiescent surroundings is valid also in flames in cross-flow. References: Battles, B. E., Seitzman, J. M., and Hanson, R. K. (1994), "Quantitative Planar Laser Induced Fluorescence Imaging of Radical Species in High Pressure Flames," AIAA 94-0229, 32nd Aerospace Sciences Meeting and Exhibit, Reno, NV. Brown, M. and Myer, T. (2002), "Laser-Induced Incandescence Measurements in the Reaction Zone of a Model Gas Turbine Combustor," 40th Aerospace Science Meeting & Exhibit, 14-17 January, 2002, Reno, Nevada, Paper no. AIAA2002-0393. Fenimore, C. P., and Jones, G. W., (1967) "Oxidation of Soot by Hydroxyl Radicals," Journal o f Physical Chemistry, Vol. 71, No. 3, pp.593-597. Goh, S. F. and Gollahalli, S. R. (2004) "Fuel Jet Dilution Effects on the Sooting Characteristic of Propylene Diffusion Flame Near Smoke Point," AIAA Conference, Reno, NV, 2004. Goh, S. F., Kusadomi, S., and Gollahalli, S. R. (2001) "Effects of Cross-Wind on Smoke Point Flow Rate on Nitrogen-Diluted Hydrocarbon Fuel Diffusion Flame," IJPGC 2001. Goh, S. F. (2003) "An Experimental and Numerical Study of Diffusion Flames in Cross-Flow and Quiescent Environment at Smoke-point Condition," Ph. D. Dissertation, School of Aerospace and Mechanical Engineering, The University of Oklahoma, Norman, OK Gollahalli, S. R., Brzutowski, T. A., and Sullivan, H. F. (1975), "Characteristic of a Turbulent Propane Diffusion Flame in a Cross-Wind," Transactions o f the Canadian Society for Mechanical Engineer, Vol. 3, No. 4, pp. 205-214. Hanson, R. K. (1986), "Combustion Diagnostics: Planer Image Techniques," Twenty-First Symposium (International) on Combustion, The Combustion Institute, PA, pp.1677-1691. Hofeldt, D. L. (1993), "Real-Time Soot Concentration Measurement Technique for Engine Exhaust Stream," SAE Paper 930079, pp. 45-57. Jenkins, T. P., Bartholomew, J. L., DeBarber, P. A., Yang, P., Seitzmen, J. M., and Howard, R. P. (2002), "Laser Induced Incandescence fir Soot Concentration Measurements in Turbine Engine Exhaust," 40th Aerospace Science Meeting & Exhibit, 14-17 January, 2002, Reno, Nevada, Paper no. AIAA2002-0828. Kelso, R. M., Lim, T. T. and Perry, A. E. (1996), "An Experimental Study of Round Jets in Cross-Flow," Journal o f Fluid Mechanics, Vol. 306, pp. 111-144. Laufer, G. (1996), Introduction to Optics and Lasers in Engineering, Cambridge University Press, New York. Mewes, B. and Seitzman, J. M. (1997), " Soot Volume Fraction and Particle Size Measurement with Laser Induced Incandescence," Applied Optics, Vol.: 33, No. 3, pp. 709-717. Shaddix, C. R. and Smyth, K. C. (1996), "Laser-Induced Incandescence Measurement of Soot Production in Steady and Flickering Methane, Propane, and Ethylene Diffusion Flames," Combustion and Flame, Vol. 107, pp. 418-452. Tamura, M., Berg, P. A., Harrington, J. E., Luque, J., Jeffries, J. B., Smith, G. P., and Crosley, D. R. (1998), "Collisional Quenching of CH(A), OH(A), and NO(A) in Low Pressure Hydrocarbon Flames," Combustion and Flame, Vol. 114, pp. 502-514. Woiki, D., Geisen, A., and Roth, P. (2000), "Time-Resolved Laser-Induced Incandescence for Soot Particle Sizing During Acetylene Pyrolysis Behind Shock Waves," Proceeding o f the Combustion Institute, Vol. 28, pp. 2531-2537. Table 1: Nominal Experimental Conditions Jet Exit Diameter Exit Velocity Exit Reynolds number Exit Froude Number Propylene Mass Flow Rate Nitrogen Mass Flow Rate Hydrogen Mass Flow Rate Cross-Flow Temperature Pressure Relative Humidity Velocity range Velocity for Diagnostic experiments Jet/Cross-flow Momentum Flux Ratio Fuel Flow Rate Nitrogen Flow Rate Cross-Flow Velocity OH Concentration Soot Concentration * - F ro m t-T e st 3.2 mm 4.2-34 m/s 520-6065 560-36760 0.0.009-0.49 kg/h 0-0.062 kg/hr 0.004-0.056 kg/hr 292 K 100.6 kpa 50% 2-4 m/s 3 m/s 0.98-58 Table 2: Experimental Uncertainties at 95% confidence level 0.06 mg/min 0.02 mg/min 0.01 m/s 2.03x10-4 mole fraction 0.02 ppm H J //" i3 '9 !♦* ■ ■ ■ 'i T 25 j '4 i m tt 4 '5 '6 '7 '3 J -* _ 9 7 '. Test section i. Flame 3. i quartz glass window 4. Contraction 5. Fuel and mixture supply 6. Single layer filter 7. Two layers filter 8. Turning ducts 9. Blowers '0. Suctionfan ''. Exhaust to atmosphere 'i . Nd-YAGlaser '3. Computer and DAQsystem '4. '5. '6. '7. '8. '9. i0. i'. ii. i3. i4. i5. ICCDcamera Camera gate controller Stanford Instrument Pulse Generator Camera imaging system Traverse mechanism Silver turning mirrors Laser Focusing lens Narrowband pass filter Beamsplitter Thermopile volume absorber detector MOPOwith FDO F igure 1: S chem atic o f LII and LIF M easurem ent Setup 0 .0 0 3 O 2.0mps o 2.5mps □ 3.0mps A 3.5mps XC4.0mps ■ | 0.0025 to ■g . 2 o' 0.002 0.0015 (A 0.001 (A T O CM 0.0005 0 0 0.002 0.004 0.006 0.008 0.01 Critical Fuel Mass Flow Rate (kg/min) Figure 2: Variation of Nitrogen Mass Flow Rate with the Fuel Mass FlowRate in Different Cross-Flow Conditions 160 140 £ 120 100 80 60 40 20 0 0% 20% 40% 60% 80% 100% Percentage of CFMFR Figure 3: Variation of Momentum Flux ratio in Different Cross-Flow Conditions Figure 4: Schematic Diagram Showing the Slice of measurement Taken at Three Flame Locations in Cross-Flow (FL= flame length) 10% & 20% CFMFR 8 7 6 5 -4 3 2 1 0 OH Mole 60% Fraction CFMFR 5.00E-3 4.75E-3 4.50E-3 4.25E-3 4.00E-3 3.75E-3 3.50E-3 3.25E-3 3.00E-3 2.75E-3 2.50E-3 2.25E-3 2.00E-3 1.75E-3 1.50E-3 1.25E-3 1.00E-3 7.50E-4 5.00E-4 2.50E-4 0.00 ■ OH Mole Fraction 5.00E-3 4.75E-3 _ 4.50E-3 4.25E-3 __ 4.00E-3 ^ 3.75E-3 1 __ 3.50E-3 I - 3.25E-3 I 3.00E-3 J 2.75E-3 I 2.50E-3 Jl 2.25E-3 2.00E-3 1.75E-3 1.50E-3 1.25E-3 1.00E-3 ZZ 7.50E-4 5.00E-4 M 2.50E-4 ■ ■ 0.00 100% CFMFR OH Mole Fraction 2.50E-3 2.38E-3 2.25E-3 2.13E-3 2.00E-3 1.88E-3 1.75E-3 1.63E-3 1.50E-3 1.38E-3 1.25E-3 1.13E-3 1.00E-3 8.75E-4 7.50E-4 6.25E-4 5.00E-4 3.75E-4 2.50E-4 1.25E-4 0.00 ^ I i i i i I i i i 11 i i i 11 ■ i i i I i i i 11 i i i i I i i i i I i i i 11 i ,1..U-l l_|_LJ_|_|_U_Lj_l_L-L-L-Lu_J_U_L±J I.JLJ_I 0 0 0 2 0 4 Y (cm) Figure 5: X-Y Plane Contours of OH Molar Concentration at 25% Flame Length in 10%, 20%, 60% and 100% CFMFR Flames at 3 m/s Cross-Flow 25% 50% 75% ppm Y (cm) F igu re 6: T om ographic P lot o f Soot C oncentration at T hree F lam e L ocations fo r 10% C F M F R F lam e at 3.0 m /s C ross-Flow F igu re 7 T om ographic P lot o f Soot C oncentration at T hree F lam e Locations for 20% C F M F R Flam e at 3.0 m /s C ross-Flow ppm X (cm) 25% FL Figure 8: 0.5 0.475 045 0.425 0.4 0.375 0.35 0.325 0.3 0.275 0.25 0.225 0.2 0.'75 0.'5 0.'25 0.' 0.075 0.05 0.025 0 ppm ppm 50% FL 0 ' Y (cm) 2 0.25 0.2375 0225 0.2'25 0.2 0. '875 0. '75 0. '625 0.'5 0. '375 0. '25 0.''25 0.' 0.0875 0.075 0.0625 0.05 0.0375 0.025 0.0'25 0 75% FL 0 ' T om ograp h ic P lot o f Soot C oncentration at T h ree F lam e L ocations for 60% C F M F R F lam e at 3.0 m /s C ross-Flow 0.25 . 0.2375 ^ 0225 0.2'25 0.2 0.'875 0.'75 0.'625 0.'5 0.'375 0.'25 0.''25 0.' 0.0875 0.075 0.0625 0.05 0.0375 0.025 0.0'25 0 25% FL 50% FL ppm 75% FL 1 0.95 0.9 0.85 0.8 0.75 0.7 0.65 0.6 0.55 0.5 0.45 0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 0 8 7 6 X (cm) 5 4 3 2 1 0 -I I I 1I I I I I I I0 1 ppm 2 Y (cm) Figure 9: Tomographic Plot of Soot Concentration at Three Flame Locations for 100% CFMFR Flame at 3.0 m/s Cross-Flow