Laser diagnostics of pulverized coal combustion in O2/N2 and O2/CO2 conditions: velocity and scalar field measurements

Exp Fluids Laser diagnostics of pulverized coal combustion in O2/N2 and O2/CO2 conditions: velocity and scalar field measurements Saravanan Balusamy 0 Simone Hochgreb 0 M. Mustafa Kamal 0 Steven M. Lowe 0 Bo Tian 0 Yi Gao 0 0 Department of Engineering, University of Cambridge , Trumpington Street, Cambridge , UK Optical diagnostic techniques are applied to a 21 kW laboratory-scale pulverized coal-methane burner to map the reaction zone during combustion, in mixtures with varying fractions of O2, N2 and CO2. Simultaneous Mie scatter and OH planar laser-induced fluorescence (PLIF) measurements have been carried out to study the effect of the oxidizer/diluent concentrations as well as the coalloading rate. The spatial distribution of soot is captured using laser-induced incandescence (LII). Additionally, velocity profiles at selected axial locations are measured using the pairwise two-dimensional laser Doppler velocimetry technique. The OH PLIF images capture the reaction zones of pilot methane-air flames and the variation of the coal flame structure under various O2/CO2 compositions. Coal particles devolatilize immediately upon crossing the flame interface, so that the Mie scatter signal almost vanishes. Increasing coal-loading rates leads to higher reaction rates and shorter flames. LII measurements show that soot is formed primarily in the wake of remaining coal particles in the product regions. Finally, differences in the mean and RMS velocity field are explained by the combined action of thermal expansion and the changes in particle diameter between reacting and non-reacting flows. 1 Introduction Coal-based thermal power plants provide almost 28 % of worlds primary energy consumption and a 36 % of all CO2 emissions (Metz et al. 2005; IEA 2012). Increasing demand for power, especially in developing countries, will lead to a further increase in CO2 emissions and other harmful pollutants. To address the problem, several carbon capture and storage (CCS) technologies have been under development. One of the keys to CCS is the ability to increase CO2 concentrations in the stream gas to make it economical for sequestration. Oxyfuel combustion, where oxygen diluted with exhaust CO2 is used as an oxidizer rather than air, offers a possible way to achieve this. Several reviews have covered experimental and numerical modeling investigations of oxyfuel coal combustion technology (Buhre et al. 2005; Kurose et al. 2009; Wall et al. 2009; Toftegaard et al. 2010; Edge et al. 2011; Chen et al. 2012). Yet there are very few quantitative measurements detailing the flame structure of such turbulent oxygen-pulverized coal flames. A number of studies have used wavelength resolved optical emission in coal combustors to obtain total radiation, typically in the infrared region, as well as an estimate of particle temperatures by using the ratio of intensities at two different wavelengths (Saito et al. 1991; Murphy and Shaddix 2006; Zhang et al. 2010; Sung et al. 2011; Draper et al. 2012; Desmira et al. 2013). Measurements of the flame structure in pulverized coal are rarer, and even fewer measurements have been attempted under realistic flame configurations. Previous measurements of pulverized coal flame structures include velocity measurements using laser Doppler velocimetry (LDV) (Pickett et al. 1999; Hwang et al. 2005, 2006a; Toporov et al. 2008; Balusamy et al. 2013), Mie scatter, OH planar laser-induced fluorescence (OH PLIF) and particle image velocimetry (PIV) by coal particles (Hwang et al. 2005, 2006a, b; Smith et al. 2002; Hayashi et al. 2013; Balusamy et al. 2013), as well as laser-induced incandescence (LII) by coal particles Hayashi et al. (2013). Most of the latter studies (Hwang et al. 2005, 2006a, b; Hayashi et al. 2013) have been made in very small laboratory diffusion flames using methane as a carrier, which demonstrated the feasibility of the diagnostics and interpretation. Other studies (Smith et al. 2002) have used Mie scatter as a qualitative index for flame structure. The present work is an advancement of the initial work carried out by Pickett et al. (1999), Hwang et al. (2005), Toporov et al. (2008) and Balusamy et al. (2013) to demonstrate how Mie scatter/OH PLIF, LDV and LII can be used in the study of the structure of swirling, turbulent, oxygenenriched pulverized coal flames. Furthermore, this work identifies under which conditions OH PLIF and Mie scatter can be usefully employed in combination with LII to determine the extent of reaction, interpret the flame structure and heat release rate. Along the way, some of the pitfalls in such imaging measurements are identified, and possible alternatives to advance the measurements are suggested. 2 Experimental methodology 2.1 Laboratoryscale coal burner Experiments are performed on an atmospheric, unconfined axisymmetric coal burner (Fig. 1). This burner has been used in previous studies, in which the flow field was characterized using LDV and PIV during the combustion of pulverized coal (Balusamy et al. 2013). The original burner design was modified to obtain higher overall reaction rates with the introduction of a central bluff body for flame stabilization (Fig. 1a). The resulting configuration consists of three coaxial tubes of 2 mm thickness, a swirler and a ceramic bluff body. Pulverized coal particles are transported to the central annulus of the burner by a mixture of metered gases. An outer pilot flame, consisting of a stoichiometric mixture of methane and air, is stabilized on the outer annulus above the axial swirler. A central flame, fueled by the incoming mixture of methane, coal, oxygen, and diluent nitrogen or carbon dioxide, is stabilized on the bluff body. The outer pilot flame and the inner coalmethane flame supply sufficient energy to ignite the coal particles and to support their reaction. A laminar co-flow stream is established in the outer annular region to isolate from external flow disturbances, providing well defined boundary conditions. The axial swirler consists of eight evenly spaced vanes of 1 mm thickness at an angle of 45 to the central axis. The swirl number, defined as the ratio of tangential to axial momentum, is estimated from the geometry (Kihm et al. 1990) as 0.77. Figure 1b depicts the experimental setup of the coal burner. A screw feeder (K-Tron, K-MV-KT20 twin screw) is used to supply the pulverized coal particles to an eductor (Schutte and Koerting). The mass flow rate of coal is Fig. 1 Schematic diagram of coal burner setup. a Coal burner top section. b Flow configuration metered by regulating the rotational speed of the DC motor driving the twin screws of the feeder. The carrier flow through the eductor entrains coal particles and transports them to the central annulus of the burner. The flow rates of the carrier air and CO2 are metered individually through a mass flow meter (Alicat, M-100SLPM-D) using a finecontrol needle valve. The flow rates of the carrier methane, c (...truncated)