(552c) Particle Image Velocimetry of Pulverized Oxy-Coal Flames

Oxy-fuel combustion of pulverized coal is a promising technology for cost-effective power production with carbon capture and sequestration that has impacts on emission reductions. To fully understand the behavior of turbulent oxy-coal flames, and to validate oxy-coal simulation models, accurate experimental data are needed on the details of the flow field and dynamics of pulverized coal flames at multiple scales. A non-intrusive laser diagnostic technique, Particle image velocimetry (PIV), has been used in the past to analyze the velocity field of flames under various combustion conditions. In this technique a laser sheet, formed using a combination of spherical and cylindrical lenses, illuminates seeding particles in the flow at two different times and images are recorded on a camera that is synchronized with the laser. Using PIV software, the images are divided into interrogation areas and correlated to determine the displacement between the two images. Knowing the time between the two laser pulses, velocity components and all velocity-derived quantities are calculated. Although PIV has seen considerable application in laboratory flames, there has been limited application of this technique in pulverized-coal flames. In this work, PIV measurements were made of pulverized oxy-coal flames in both a laboratory-scale laminar diffusion flame burner and a 100 KW turbulent diffusion flame burner. Significant method development was carried out using the laboratory-scale burner setup, prior to application of the technique to the 100 KW test facility, and considerable effort was expended in the design of a stable lab-scale pulverized coal burner that could accommodate the PIV measurements. The lab-scale burner utilizes four concentric streams to control the ignition and flame shape. In this design, the coal is transported from the coal feeder by a carrier gas (air) through the central tube of the burner. Immediately circling the central coal jet is a natural gas pilot whose purpose is to ignite the coal. Outside of the pilot ring is an air annulus that will provide air to both the pilot and coal flames and will allow for control of the overall stoichiometric ratio during coal combustion. The final annular ring (outside of the air stream) is a pre-mixed natural gas/air stream. The primary purposes of this stream are to heat the air stream, facilitating ignition of the coal jet, and to reduce heat loss from the flame. The design concept for the coal feeder is a motor-driven unit that constantly advances a powdered coal surface toward a jet of air that fluidizes the coal and carries it out of the feeder to the central tube of the burner. A vibration unit is used to disrupt the coal and facilitate the coal fluidization. Preliminary results with the laboratory-scale burner show that the optimal mass flow rate of coal was in the range from 30 mg/min to 2000 mg/min, depending on the conditions used. Coal flow rate values lower than 250 mg/min were preferred for PIV tests, since for these conditions the PIV images permitted the isolation of individual coal particles in the flame, facilitating the estimation of particle size and particle velocities. PIV images obtained for coal flow rates larger than 300 mg/min usually showed a cloud of particles in the flame, making it difficult to isolate individual particles. Coal particle velocities were obtained under different global equivalence ratios of the system (ranging from 0.45 to 1.1) for a selected condition of mass flow rate of coal (150 mg/min). For this coal flow rate, it was observed that as the amount of air decreases (higher equivalence ratio) the particle velocity decreases. The average particle velocity varied from 1.25 m/s for a 0.45 equivalence ratio to 0.3 m/s for a 1.1 equivalence ratio. This observation is consistent with an overall reduction in volumetric flow through the burner, resulting in lower particle velocities for the majority of the flame, even though the initial coal particle jet velocity remained unchanged. Different sizes of particles are present in the flame, with larger sizes concentrating around the center. As the flow field contains different sizes of coal particles, it is useful to try and isolate different particle sizes and compare the velocity field for different coal sizes. In addition, the vertical velocity (the y-component of the velocity) and radial velocity, as well as coal-particle size distribution will be investigated at different heights above the burner for the coal-based flame under different equivalence ratios to account for the changes in the particle size distribution in the near- and far-burner regions. These same methodologies are currently being applied to the 100 KW turbulent flame, and the results of these trials will also be presented.