Energy Fuels 2010, 24, 346–354 Published on Web 09/15/2009
: DOI:10.1021/ef900836a
Influence of Different Outer Secondary Air Vane Angles on Flow and Combustion Characteristics and NOx Emissions of a New Swirl Coal Burner Jianping Jing, Zhengqi Li,* Guangkui Liu, Zhichao Chen, and Feng Ren School of Energy Science and Engineering, Harbin Institute of Technology, 92, West Dazhi Street, Harbin 150001, People’s Republic of China Received July 31, 2009. Revised Manuscript Received September 1, 2009
Measurements were taken for a 300 MWe wall-fired pulverized-coal utility boiler installed with eight centrally fuel rich swirl coal combustion burners in the bottom row of the furnace during experiments. For various outer secondary air vane angle settings, flow characteristics, gas temperature and gas species concentrations in the burner region were measured. The results show that with decreasing outer secondary air vane angle, the swirl intensity of air, divergence angles and maximum length and diameter of the central recirculation zone all increased; the turbulence intensity of the jet flow first increased and then quickly decreased. With decreasing outer secondary air vane angle, the rate of increase in the gas temperature in the early stage and the rate of decrease in the later stage increased along the jet flow direction. For outer secondary air vane angle of 25°, the O2 concentration along the jet flow direction in the burner region was low, and the CO concentration decreased with increasing outer secondary air vane angle. For outer secondary air vane angles of 30° and 35°, the NOx concentrations were less than those for 25° and 40° outer secondary air vane angles. Gas species concentrations in the side wall region varied slightly with a change in the outer secondary air vane angle.
front-wall-fired, pulverized-coal utility boiler.3-5 Vikhansky et al. measured heat fluxes in a 550 MWe, opposite-wall-fired, pulverized-coal utility boiler.6 Experiments have been performed in pulverized-coal, tangentially fired, down-fired, and wall-fired boilers,7-16 but few detailed measurements have been taken in the burner region. Different aerodynamic fields lead to variations in the coal combustion process. To analyze the influence of aerodynamic behavior on pulverized-coal combustion, researchers have studied the aerodynamic behavior of full-scale furnaces by performing experiments with small-scale models.17,18
1. Introduction Coal-fired plants account for a great proportion of the energy capacity in the international energy production and consumption structure, with the power industry being one of its major consuming industries. However, of prime concern in the operation of thermal power units is the ever-present problem of high NOx emissions, the control of which has become increasingly strict globally in recent years. The objective of the Cost Abatement for Effective NOx Reduction in PF Coal-Fired Power Plants (CAFENOX) project is to check the ability of primary techniques to attain exhaust NOx emission levels imposed by the European Union. From 2008, the allowed concentrations for power plants rated over 50 MWe will be set at 500 mg of NO2/Nm3 at 6% O2. From 2016, the limit for power plants rated over 500 MWe will be 200 mg of NO2/Nm3 at 6% O2.1 In China, the permissible NOx emission limit was set at 450 mg/m3 for power plants built from 2004 onward. Intensive research has been conducted around the world to reduce pollutant emissions generated in the combustion zone. Li et al. proposed a new low-NOx pulverized-coal burner technology, the centrally fuel-rich (CFR) swirl coal combustion burner, based on a radial-bias combustion burner and an enhanced ignition-dual register burner.2 Industrial experiments performed on full-scale boilers have revealed coal combustion characteristics and the mechanism of NOx formation. Costa et al. measured local mean gas species concentrations (O2, CO, CO2, and NOx), gas temperatures, and char burnout at several ports in a 300 MWe,
(3) Costa, M.; Silva, P.; Azevedo, J. L. T.; Carvalho, M. G. Combust. Sci. Technol. 1997, 129, 277–293. (4) Costa, M.; Silva, P.; Azevedo, J. L. T. Combust. Sci. Technol. 2003, 175, 271–289. (5) Costa, M.; Azevedo, J. L. T. Combust. Sci. Technol. 2007, 179, 1923–1935. (6) Vikhansky, A.; Bar-Ziv, E.; Chudnovsky, B.; Talanker, A.; Eddings, E.; Sarofim, A. Int. J. Energy Res. 2004, 28, 391–401. (7) Black, D. L.; McQuay, M. Q. Combust. Fire 1996, 328, 19–27. (8) Xue, S.; Hui, S. E.; Liu, T. S.; Zhou, Q. L.; Xu, T. M.; Hu, H. L. Fuel Process. Technol. 2009, 90, 1142–1147. (9) Xue, S.; Hui, S. E.; Zhou, Q. L.; Xu, T. M. Energy Fuels 2009, 23, 3558–3564. (10) Zhou, Q. L.; Zhao, Q. X.; Li, N.; Chen, X.; Xu, T. M.; Hui, S. E. Int. J. Energy Res. 2009, 33, 235–254. (11) Butler, B. W.; Wilson, T.; Webb, B. W. Proc. Combust. Inst. 1992, 24, 1333–1339. (12) Ren, F.; Li, Z. Q.; Zhang, Y. B.; Sun, S. Z.; Zhang, X. H.; Chen, Z. C. Energy Fuels 2007, 21, 668–676. (13) Fan, J. R.; Sun, P.; Zheng, Y. Q.; Ma, Y. L.; Cen, K. F. Fuel 1999, 78, 1387–1394. (14) Li, Z. Q.; Yang, L. B.; Qiu, P. H.; Sun, R.; Chen, L. Z.; Sun, S. Z. Int. J. Energy Res. 2004, 28, 511–520. (15) Tree, D. R.; Webb, B. W. Fuel 1997, 76, 1057–1066. (16) Queiroz, M.; Bonin, M. P.; Shirolkar, J. S.; Dawson, R. W. Energy Fuels 1993, 7, 842–851. (17) Li, Z. Q.; Chen, Z. C.; Sun, R.; Wu, S. H. J. Energy Inst. 2007, 80, 123–130. (18) Dixon, T. F.; Truelove, J. S.; Wall, T. F. J. Fluids Eng. 1983, 105, 197–203.
*To whom correspondence should be addressed. Telephone: þ86451-86418854. Fax: þ86-451-86412528. E-mail:
[email protected]. (1) Thomas, L. B.; Francisco, C.; Sebastien, C.; Stanislas, P.; Jacques, B.; Bernard, B. Fuel 2007, 86, 2213–2220. (2) Li, Z. Q.; Jing, J. P.; Chen, Z. C.; Ren, F.; Xu, B.; Wei, H. D.; Ge, Z. H. Combust. Sci. Technol. 2008, 180, 1–25. r 2009 American Chemical Society
346
pubs.acs.org/EF
Energy Fuels 2010, 24, 346–354
: DOI:10.1021/ef900836a
Jing et al.
Figure 1. Schematic view of the boiler with the burners (all dimensions in meters).
In the present work, cold air-flow experiments were carried out for a small-scale burner model and in situ experiments were carried out for a CFR burner. Measurements were conducted for a 300 MWe wall-fired pulverized-coal utility boiler. The measurement was performed in No. 4 burner in the bottom row on the rear wall of the furnace. Local mean concentrations of O2, CO, and NOx and gas temperatures were recorded at positions within the CFR burner for different outer secondary air (OSA) vane angles. 2. Utility Boiler A B&W B-1025/16.8 M boiler with a 300 MWe unit was made by Babcock and Wilcox Beijing Co. Ltd. The oppositewall-fired, pulverized-coal boiler with a dry-ash-type furnace is equipped with 20 enhanced ignition-dual register burners, of which 12 are arranged in three rows on the front wall of the furnace. The other 8 burners are arranged in two rows on the rear wall, opposite the 8 burners in the top and bottom rows on the front wall. Figure 1 shows the schematic view of the boiler with the burners. Five medium-speed mills and a positive-pressure direct-fired system are used to supply pulverized coal to the burners. To increase the combustion stability of the boiler, 8 enhanced ignition-dual register burners in the bottom row of the furnace were replaced with CFR burners. Figure 2 is a schematic diagram of a CFR burner with 16 bent-shaft vanes in the inner secondary air duct and 12 tangential vanes at the entrance of the outer secondary air duct. In comparison to the enhanced ignitiondual register burner, the CFR burner has cone separators instead of a particle deflector and conical diffuser in the primary air tube to affect the distribution of pulverized coal in the primary air. Under the influence of the cone separators, pulverized coal carried by primary air is concentrated in the central zone of primary air, which results in a coal-rich flow within this zone and a coal-lean flow peripheral to it. Experiments were carried out for the utility boiler to study the flow characteristics, combustion characteristics, and NOx emissions for different OSA vane angle settings within the
Figure 2. Cross-section of the CFR burner and position of the monitoring pipe (dimensions in meters): (1) primary air duct, (2) monitoring pipe, (3) cone separators, (4) radial vanes, (5) inner secondary air duct, (6) tangential vanes, and (7) outer secondary air duct.
CFR swirl burners. Table 1 lists the design parameters for the burner. 3. Cold Air Experiments of the CFR Burner Model Using isothermal modeling technology, a cold experiment was carried out in the laboratory using a CFR burner model one-quarter of the size of the prototype. The test facility is illustrated in Figure 3, where x is the distance to the exit of the burner along the jet flow direction, r is the distance to the chamber axis along the radius direction, and d is the diameter of the outer secondary air duct (d=0.374 m). An IFA300 constant temperature anemometer system was used to measure the air velocity at various measurement sites. Using a probe with hot-film sensors, we measured the three-dimensional 347
Energy Fuels 2010, 24, 346–354
: DOI:10.1021/ef900836a
Jing et al.
Figure 3. Cold air experiment system.
Figure 4. Mean axial velocity profiles for different OSA vane angles.
primary air-flow zone, and the peak near the wall defines the secondary air-flow zone. The peak values near the wall increase with decreasing OSA vane angles. Because of the higher velocity gradients between primary and secondary air, mixing between them develops quickly, owing to the high mass-transfer rate. For cross-sections at x/d g 0.5, peak values of the primary air-flow zone have already vanished because primary air has already mixed with secondary air. When the OSA vane angles are small, the peak value of axial velocities within the secondary air-flow zone is high and primary and secondary air mix quickly. The larger the OSA vane angle, the weaker the swirl intensity of the exit flow and the smaller the centrifugal force of fluid micelles. Consequently, the air flow is more centralized in the jet flow center; the jet flow expansion angle reduces; and the recirculation zone reduces accordingly. As shown in Figure 5, for four typical cross-sections from the exit of the burner jet to a distance ratio of x/d = 0.25, the radial mean velocity distributions are double-peak structures; the inner peak near the burner center determines the primary
Table 1. Design Parameters of the CFR Burner in the Utility Boiler quantity exit area of the primary air (m2) exit area of the inner secondary air (m2) exit area of the outer secondary air (m2) temperature of the primary air (°C) temperature of the secondary air (°C) mass flow rate of the primary air (kg s-1) mass flow rate of the inner secondary air (kg s-1) mass flow rate of the outer secondary air (kg s-1)
CFR burner 0.2597 0.4979 0.6677 75 353 5.80 3.59 8.37
flow field at the exit of the CFR burner.19 The error in velocity measurements was less than 5%. Figures 4-6, respectively, show the profiles of the axial, radial, and tangential mean velocities, with vane angles set at 25°, 30°, 35°, and 40°. It can be seen in Figure 4 that, from the burner jet (x/d = 0) to the x/d = 0.25 cross-section, there are two peaks in the profiles of axial mean velocities along the radius direction: the peak near the burner center defines the (19) Mueller, U. R. J. Fluid Mech. 1982, 119, 155–172.
348
Energy Fuels 2010, 24, 346–354
: DOI:10.1021/ef900836a
Jing et al.
Figure 5. Mean radial velocity profiles for different OSA vane angles.
Figure 6. Mean tangential velocity profiles for different OSA vane angles.
air-flow zone, and the outer peak determines the outer secondary air-flow zone. The maximum value of the outer peak is always higher than that near the center; meanwhile, the maximum value of the outer peak is higher for smaller OSA vane angles. In conjunction with primary air diffusing into the secondary air flow, secondary air flow diffuses and the two peak values of the radial velocity decay gradually, while their positions move closer to the wall. With further jet flow development, the radial velocities flatten out and the differences in the mean radial velocities are very small for the four cross-sections presented after x/d = 1.0. From the burner jet exit to the x/d = 0.5 cross-section, the radial velocities of the CFR burner along the chamber axis always easily attain negative values under different OSA vane angles. This shows the centerline migration of primary air, which is advantageous in increasing pulverized coal concentrations within the burner center region.
Figure 6 shows that for four different vane angles, the tangential mean velocities have a single peak, the maximum value of which increases with a decreasing OSA vane angle. Because primary air is nonswirling and the recirculation zone is mainly axial backflow, the tangential velocity in the central recirculation zone and at the primary air outlet is low. The swirl vane in the outer secondary air primary flow region forms a Rankine vortex with high tangential velocity; the central narrow zone is a combination of a solid-body rational core and the large peripheral vortex-free zone. The peak tangential velocity at the outer secondary air outlet increases with a decreasing vane angle, while the flow swirl intensity increases. The tangential velocity decays faster for a smaller vane angle because the mixing between flows is strong. Meanwhile, the decay of the tangential velocity is faster than the decay of the axial velocity. Thus, the tangential swirl 349
Energy Fuels 2010, 24, 346–354
: DOI:10.1021/ef900836a
Jing et al.
Figure 7. Turbulence intensity profiles for different OSA vane angles.
characteristic disappears rapidly in a strong swirl jet because of turbulence mixing effects, as seen in cross-sections at x/d g 1.0 when the tangential velocity is low, and hence, the axial flow is the main mode of flow. Figure 7 shows the changes in the turbulence intensity corresponding to the four vane angle settings. The turbulence intensity T is defined as qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi U 0 2 þV 0 2 þW 0 2 ð1Þ T ¼ U0
Table 2. Cold Air-Flow Experiment Parameters of the CFR Burner mass flow rate of primary air (kg s-1)
mass flow rate of inner secondary air (kg s-1)
mass flow rate of outer secondary air (kg s-1)
4.38
3.12
7.40
measurements was 0.1 m. We estimated the uncertainty in establishing the location of the central recirculation zone border to be of the order of 0.1 m. From the flow direction of the cloth, the jet borders and the central recirculation zone boundary of the burner were determined.17 Figure 8 shows three aerodynamic field profiles corresponding to OSA vane angles of 25°, 30°, and 35°, where L is the distance to the burner exit along the jet flow direction, r is the distance to the chamber axis along the radial direction, and D is the burner outer diameter of the outer secondary air duct (D=1.495 m). The figure shows that there is a distance between the centers of the central recirculation zone and the burner port, for which slagging around the burner port can effectively be prevented from occurring depending upon the OSA vane angle. When the OSA vane angle varies from 35° to 25°, both the length and diameter of the recirculation zone increase markedly, owing to the increase in swirl intensity of the secondary air with a decreasing OSA vane angle. With an OSA vane angle of 30°, the length of the recirculation zone was 0.94D, while the maximum diameter of the recirculation zone was 0.94D. With an angle of 35°, the length and maximum diameter of the recirculation zone were 0.67D and 0.74D, respectively. In comparison to recirculation zones for OSA vane angles of 30° and 35°, the zone was larger when the angle was 25°, with both the length and maximum diameter of the recirculation zone being 1.2D. Because the distance between the central lines of two adjacent burners in the furnace was 1.66D, a larger recirculation zone and an extended vane angle might affect the interaction between two adjacent burners, thus causing instabilities within the flame. At the same time, a larger recirculation zone might advance the ignition of pulverized coal, potentially destroying the burner port. During operation, a larger recirculation zone did not provide better performance, and therefore, we chose a suitable OSA vane angle.
where U0 , V0 , and W0 are the axial, radial, and tangential velocity fluctuations, respectively, and U0 is the mean velocity of the burner jet. The figure shows that the distribution of the turbulence intensity in the flow field is not even. Values are small in the recirculation zone and at the boundary of the jet. The boundary of the recirculation zone and main flow zone of secondary air has high turbulence. The turbulence intensity is not a maximum at the burner nozzle (x/d = 0). As the air spreads to areas downstream, turbulence energy is generated continually. The turbulence intensity peaks in the zone bounded by cross-sections at x/d = 0.25 and 0.5, after which it gradually decreases. Thus, regions with large turbulence intensities are where pulverized coal burns strongly. In practical operation, the ignition distance of the air/coal flow should be controlled in this area. When vane angles are small, the turbulence intensity at the outlet is at a high level and turbulent mixing is strong, leading to the early ignition of pulverized coal. Because the early mixing of the flow is strong, the dissipation of turbulent energy quickens. In the zone bounded by x/d = 1.5 and 2.5, the turbulence intensity decreases with a decreasing vane angle and increases for small vane angles. 4. In Situ Industrial Cold Flow Experiments Characteristics of the flame can be analyzed and predicted from in situ industrial cold air-flow experiments. Table 2 lists experimental parameters. In these experiments, a coordinate frame was set at the burner outlet and a thin cloth was fixed for each grid of the frame. The transverse distance between two 350
Energy Fuels 2010, 24, 346–354
: DOI:10.1021/ef900836a
Jing et al.
Figure 10. Water-cooled stainless-steel probe. Table 3. Characteristics of the Coal Used in the Experiments Proximate Analysis (as Received, wt %)
Figure 8. Profiles of the jet border and central recirculation zone boundary of the CFR burner for different OSA vane angles.
volatile matter
ash
33.15
27.13
moisture
fixed carbon
net heating value (kJ/kg)
11.8
40.82
17790
Ultimate Analysis (as Received, wt %) carbon
hydrogen
sulfur
nitrogen
oxygen
48.05
2.51
1.23
0.54
8.74
Table 4. Boiler Design and Operation Parameters quantity flow rate of the main steam (tons/h) pressure of the main steam (MPa) temperature of the main steam (°C) reheat steam outlet temperature (°C) reheat steam outlet pressure (MPa) coal feed rate (tons/h) primary air flow rate (tons/h) secondary air flow rate (tons/h) primary air temperature (°C) secondary air temperature (°C)
operation parameter
design parameter
974.3 16.7 537.8 538.5 3.2 148.1 251.7 663.6 73.0 362.0
1025.0 16.8 540.0 540.0 3.4 154.2 285.3 731.3 75.0 353.0
Gases were sampled using a water-cooled stainless-steel probe, shown in Figure 10, and analyzed online on a Testo 350 M instrument. Gas temperatures were measured using a nickel chromium-nickel silicon thermocouple placed inside a water-cooled stainless-steel probe composed of a water-in pipe, water-out pipe, sampling pipe, and outer pipe. The high-pressure cooling water coming from the water-in pipe cools the sampling pipe and then flows from the water-out pipe. The gas is sampled by a sampling pipe, where upon smoke entering, temperatures decease rapidly and pulverized coal stops burning. The samples pass through filtration devices into a Testo 350 M gas analyzer. The end of the bare thermocouple was exposed in the furnace, and therefore, the temperature measured should be higher than the local gas temperature because of the high flame radiation. Meanwhile, because of radiation passing between the bare thermocouple and the water-cooled wall, which are in close proximity, the temperature measured should be lower than the local gas temperature. Considering these two effects, the measurement error should be less than 50 °C. The accuracy of the Testo 350 M gas analyzer is 1% for O2 and 5 ppm for CO, NO, and NO2. Each sensor was calibrated before measurement. During the experiments, the utility boiler operated stably with a full load. Table 3 lists the characteristics of the coal used in the experiments. Table 4 summarizes the design and operation parameters of the boiler. The values given in Table 4 are averages over the duration of the experiments. 5.2. Results and Discussion. Figure 11 shows profiles of the gas temperature of the CFR burner for different OSA vane angles. Figure 11a shows that all gas temperatures increase
Figure 9. Schematic top view of the burners and the monitoring ports at the bottom of the furnace (dimensions are in meters).
5. Measurements of the Gas Temperature and Gas Species Concentrations in the Burner Region 5.1. Data Acquisition Techniques. Because a swirl burner works independently, there is little variation in the temperature and velocity field. Data were obtained for gas temperature and gas species concentrations in the region of the No. 4 burner located at the bottom of the rear wall (see Figure 9). Measurements were obtained at positions along the x direction near the burner region using a water-cooled stainlesssteel probe inserted through monitoring ports (see Figure 2) in the burner. The initial measurement position along the x direction was on the rear wall. Measurements at positions along the direction from the side wall to the burner were taken through a monitoring port in the side wall, as shown in Figure 9. To prevent the water-cooled stainless-steel probe from being distorted by hot gas, the distance between the measurement point and the side wall was confined to be less than 1.6 m. Thus, data do not represent the center of the burner. Studies of the center of the burner are performed using numerical simulations. 351
Energy Fuels 2010, 24, 346–354
: DOI:10.1021/ef900836a
Jing et al.
Figure 11. Profiles of the gas temperature measured for different OSA vane angles (a) along the x direction and (b) in the radial direction near the burner.
Figure 12. Profiles of the O2 concentration measured for different OSA vane angles (a) along the x direction and (b) in the radial direction near the burner.
initially then decrease along the jet direction, with high rates of increase in the early stage. The smaller the angle, the quicker the temperature rises in the early stage and the faster it decays in the later stage. Gas temperatures increase sharply and then remain at a high level at a distance from the burner, thus achieving rapid combustion of the pulverized coal in the high-temperature gas. At positions away from the burner, gas temperatures gradually decrease, owing to fuel consumption and mixing of the primary and secondary air flows. From Figures 7 and 8, we see that the swirl intensity of secondary air increases with a decreasing OSA vane angle and the central recirculation zone begins to enlarge, thereby drawing in more high-temperature gas and burning off pulverized coal in advance. There is a high heating rate and high gas temperature in the initial stage. However, because of the large quantity of pulverized coal burnt early and, thus, weakening the mixing of pulverized coal and secondary air, gas temperatures decrease quickly. As observed in Figure 11b, the gas temperature increases from the side wall to the burner center. Thus, gas close to the high-temperature central recirculation zone is at a higher temperature than that near the water-cooled side wall. Figure 12 shows O2 concentration profiles of the CFR burner for different OSA vane angles. Figure 12a shows that
O2 concentrations for all OSA vane angles decrease sharply at first and then increase slowly later along the jet direction. O2 concentrations are at their lowest when the OSA vane angle is set to 25°. The reason for the initial sharp decrease in the O2 concentration along the x direction is that the pulverized coal combusts rapidly and consumes a great deal of oxygen. At positions away from the rear wall, O2 concentrations increase because oxygen in the secondary air is supplied to the primary air. When the OSA vane angle is 25°, the swirl intensity of the flow is strong, the mixing of primary air and secondary air strengthens, and pulverized coal burns early, consuming large amounts of oxygen. Meanwhile, O2 concentrations extend to the side wall mainly with the swirl of secondary air. Thus, O2 concentrations are low in the burner central zone. As observed in Figure 12b, O2 concentrations for various vane angles change slightly from the side wall to the burner. O2 concentrations are highest near the side wall when the vane angle is 25° and lowest when the angle is 35° but still above 9%. The CFR swirl coal combustion burner having high O2 concentrations for different OSA vane angles near the side wall is beneficial in avoiding hightemperature corrosion near the side wall.20 (20) You, C. F.; Zhou, Y. Energy Fuels 2006, 20, 1855–1861.
352
Energy Fuels 2010, 24, 346–354
: DOI:10.1021/ef900836a
Jing et al.
Figure 13. Profiles of the CO concentration measured for different OSA vane angles (a) along the x direction and (b) in the radial direction near the burner.
Figure 14. Profiles of the NOx concentration measured for different OSA vane angles (a) along the x direction and (b) in the radial direction near the burner.
Figure 13 shows CO concentration profiles of the CFR burner for different OSA vane angles. On the one hand, Figure 13a indicates that CO concentrations along the x direction initially increase and then decrease. With an increasing O2 concentration, the CO concentration decreases along the x direction. When pulverized coal begins to burn, it does so quickly, expending a large amount of O2. At this point, there is little secondary air mixing with primary air, which results in large CO concentrations without sufficient O2. With increasing distance between the measurement point and water-cooled wall, more secondary air mixes with primary air and CO transforms into CO2. On the other hand, CO concentrations decrease with an increasing vane angle. The swirl intensity of the flow is strong for smaller angles, while the larger the central recirculation zone, the stronger the mixing of primary air and secondary air. In addition, the earlier pulverized coal is burnt, the lower the O2 concentration. It is easy to form a zone of high temperature and high concentration. As observed in Figure 13b, CO concentrations for different vane angles are low and almost constant over the region from the side wall to the burner. Figure 14 shows the NOx concentration distributions for different OSA vane angles. From Figure 14a, all NOx concentrations go through a sequence of rising, falling, rising
again, and finally becoming constant. When the angle of the OSA vane is set at 25°, the NOx concentration remains high. When the angle is set to 40°, the NOx concentration is initially low and then increases. For angles of 30° and 35°, NOx concentrations are relatively low in the central zone of the burner. For all four settings, NOx concentrations peak near the burner nozzle. This is mainly due to pulverized coal burning very quickly and NOx concentrations increasing. Pulverized coal and volatile concentrations are high in the central recirculation zone, but there is little O2. In this reducing environment, NOx formation for all four vane settings is low and most of the reactive nitrogen is converted to N2.21 With the continual supply of secondary air and further char combustion, the NOx concentration has a second peak value for each vane setting. Finally, with char burnout, NOx concentrations are essentially constant. For a vane angle set at 25°, the NOx concentration is obviously higher in the early combustion stage than it is for the other three settings. This is because the central recirculation zone is large for an angle of 25°. Primary air and secondary air mix strongly in the early stage, and the turbulence intensity is (21) van der Lans, R. P.; Glarborg, P.; Dam-Johansen, K. Prog. Energy Combust. Sci. 1997, 23, 349–377.
353
Energy Fuels 2010, 24, 346–354
: DOI:10.1021/ef900836a
Jing et al.
high. Pulverized coal ignites early, and temperatures are high; thus, NOx production is high. With a vane angle of 40°, the central recirculation zone is small. The mixing of primary and secondary air is weak, and pulverized coal ignites late. Thus, the NOx concentration is a little lower than that for the other three settings. However, in late combustion, a large portion of pulverized coal that was not burnt in the early stage begins to ignite. Thus, the NOx concentration rises quickly and is higher than that for the other three angle settings. Figure 14b shows that, along the radial direction, the NOx concentration for each setting rises in an undulating manner as the burner core is approached.
decreasing OSA vane angle, the rate of increase in the gas temperature in the early stage and the rate of decrease in the later stage increased along the jet flow direction. For an OSA vane angle of 25°, the O2 concentration along the jet flow direction in the burner region was low and the CO concentration decreased with an increasing OSA vane angle. For OSA vane angles of 30° and 35°, the NOx concentrations were less than those for angles of 25° and 40°. (3) For different OSA vane angles, the gas temperatures and gas species concentrations varied slightly in the side-wall region. High O2 concentrations and low CO concentrations of the CFR burner in the side-wall region could inhibit high-temperature corrosion and slagging on the water wall. (4) For an OSA vane angle of 35°, a more appropriate central recirculation zone was established with better combustion characteristics and lower NOx emissions.
6. Conclusions The flow and combustion characteristics of a CFR swirl coal combustion burner were studied for different OSA vane angles in laboratory experiments and in situ experiments. The results are as follows: (1) With a decreasing OSA vane angle, the swirl intensity of the gas, the divergence angle, and the maximum length and diameter of the central recirculation zone all increased. The turbulence intensity of the jet flow increased initially and then quickly decreased. (2) With a
Acknowledgment. This work was supported by the Hi-Tech Research and Development Program of China (Contract 2007AA05Z301), Key Project of the National 11th 5-Year Research Program of China (Contract 2006BAA01B01), Heilongjiang Province via 2005 Key Projects (Contract GC05A314), and the Postdoctoral Foundation of Heilongjiang Province (LRB07-216).
354
