Flame pattern analysis for $60{\mathrm{kW}}_{\mathrm{th}}$ flames under conventional air-fired and oxy-fuel conditions for two different types of coal

Abstract

The present work is dedicated to the experimental investigation of the influence of fuel-type and oxidizer composition upon flame structure and behaviour of swirl-stabilized pulverized coal flames. Detailed flame measurements are conducted by employing a combination of flame-intrusive and non-intrusive measurement techniques which can provide complementary data about flow fields, major product species and radiative heat transfer from the flames. Four flames with constant thermal output ($60{\mathrm{kW}}_{\mathrm{th}}$) and stoichiometry are employed. While previous studies of the same configuration were limited to one fuel (Rhenish lignite), the influence of fuel type is investigated here by additionally measuring the same set of parameter for Prosper Haniel bituminous coal. Two reactive atmospheres (conventional air and oxy-fuel with an ${\mathrm{O}}_2/{\mathrm{CO}}_2$ ratio of $25/75\mathrm{vol}\%$) are employed to investigate the impact of changes in oxidizer. The combined analysis of measurement results show that flame length is predominantly controlled by the effective swirl intensity when the flame ignites and stabilizes in the vicinity of the burner. Further on, measurements from narrow-band flame imaging and heat flux measurements show that the location of peak combustion intensity is determined by the flow inlet conditions (at the burner). This being key parameters, that could be employed to match heat transfer profiles when transitioning from conventional air firing to oxy-fuel in existing power plants.

Introduction

Despite a decline in use of coal for electricity production over the last years, the overall consumption is projected to rise for the coming decades, based on the current energy policies [1]. Scenarios projecting the ${\mathrm{CO}}_2$-release from coal power plants, remain far below the ${\mathrm{CO}}_2$-emission-reduction required to meet the ambitious goals of the Intergovernmental Panel on Climate Change (IPCC) [2], necessary to prevent an increase of global mean temperature by more than $1.5°\mathrm{C}$ by mid-21st century. The sustainable development scenario projected in the World Energy Outlook (WEO) from 2018 [1] requires a decrease of ${\mathrm{CO}}_2$ emissions from use of coal by a factor of ten compared to the currently stated world-wide policies. Therefore different techniques to reduce carbon dioxide emissions have to be implemented. While the WEO highlights the need for Carbon Capture Utilization and Storage (CCUS) techniques, it emphasizes the lack of progress in this area and the need of retrofitting or equipping new power plants with CCUS techniques in order to meet the required emission reductions. This does not only require political action, but also scientific development in this area, to make existing techniques like oxy-fuel combustion more mature. Oxy-fuel combustion, proposed by Abraham et al. [3] in 1982 is a method where the flue gas is enriched with ${\mathrm{O}}_2$ and fed as oxidizer to the combustion process. The resulting flue gas consists primarily of ${\mathrm{CO}}_2$, simplifying its sequestration. Therefore the differences in combustion, due to the different oxidizer composition [4], have to be examined, i.e. the influence of changing the oxidizer’s composition from the conventional ${\mathrm{O}}_2/{\mathrm{N}}_2$ mixture with respective volume fractions of $21\mathrm{vol}.\%/79\mathrm{vol}.\%$ to an ${\mathrm{O}}_2/{\mathrm{CO}}_2$ composition in oxy-fuel-combustion with different volume fractions. Differences may for example arise from changes of heat capacity or gas radiation [4]. Thus, for retrofitting existing coal power plants, comprehensive studies and models are needed to account for changes in fuel type or oxidizer composition on flame stability and heat transfer characteristics. Comprehensive reviews on oxy-fuel combustion are given in [4], [5].

Investigations of major species concentration have been carried out by H. Liu et al. [6] in a 20 kW combustor finding a decrease in NOx production when changing from air to oxyfuel combustion, while SOx production was not influenced by changes in oxidizer composition. K. Andersson et al. [7] found a decrease of $70\%$ when changing from air to oxyfuel combustion with flue gas recycling in a $100{\mathrm{kW}}_{\mathrm{th}}$ test facility. However, they show that the conversion of fuel-N to NO can be even higher during oxyfuel combustion, in agreement with previous studies, where NO mainly reacts to molecular ${\mathrm{N}}_2$ by chemical processes in the reaction zone [8].

The flow field of pulverized coal flames has been mostly investigated by means of Laser Doppler Anemometry (LDA), determining the coal particle velocity. Different configurations have been considered, covering a range from several tens of kilowatt to a half MW size, e.g. Schnell et al. [9], [10] over the MW-range, e.g. [11], [12], [13] to several tens of MW, e.g. [14]. Corrêa da Silva et al. [15] have shown, that the stabilization of oxycoal flames is determined in the burner front and strongly depends on the inflow conditions. Similar observations have been made by Leuckel and Fricker [16], postulating from velocity measurements below the burner that flow stabilization of IFRF Type-2 flames occurs within the burner quarl. In a recent study, Becker et al. [17] employed a burner geometrically similar to the one used in the present study, but with optical access to the quarl (diffuser). The burner was operated under air and oxy-fuel conditions with methane as a fuel. This study shows that the flame dynamics occurring inside the quarl, a location usually not accessible by instrumentation, are important to understand the flame stabilization mechanisms. Therefore, information about the flow fields near the burner is necessary in order to understand flame stabilization behavior.

The radiative heat flux characteristics of $100{\mathrm{kW}}_{\mathrm{th}}$ lignite-flames with varying ${\mathrm{O}}_2$-content have been investigated in [18], finding similar radiative heat flux characteristics at $25\mathrm{vol}.\%{\mathrm{O}}_2$-content in the ${\mathrm{O}}_2$/${\mathrm{CO}}_2$ mixture as compared to conventional air combustion. The influence of coal type on radiative heat flux at different ${\mathrm{O}}_2$-content in the recycled ${\mathrm{O}}_2/{\mathrm{CO}}_2$ atmosphere has been evaluated by J.P. Smart et al. [19], concluding that the ${\mathrm{O}}_2$-content at which heat fluxes of oxy-fuel combustion match those resulting from combustion in conventional air-atmosphere are coal-type dependent. Hjärtstam et al. [20] performed flue gas sampling for lignite fired flames under air- and different oxy-fuel conditions, finding that NO production goes down as flue gas is recycled and that CO-emissions are not influenced by the change in oxidizer even though the spatial distribution of CO production inside the flames changes with oxidizer composition.

In recent publications [21], [22], [23], [24], [25], flow field, local gas composition, particle temperature measurements or narrow-band imaging have been performed on the same setup as used in the present study. The focus in terms of fuel-type has been put on lignite and experiments were dedicated to investigate the influence of oxidizer composition. Various techniques have been employed and improved for detailed flow [24], [26] or temperature measurements [25] and gas sampling [22] was applied to map major product species. It was shown that due to density changes of the oxidizer, when changing its composition, the associated momentum changes at constant inlet velocities are mainly responsible for changes in flow field and flame structure [23]. Therefore the momentum of the inlet streams could be identified to be an important tuning parameter for the flame structure when switching from air- to oxy-fuel-atmosphere.

In [23] it was shown that for the same configuration as in the present study, flames with same momentum have similar flame shape, a principle that can be used for conversion of existing conventionally air fired power plants to oxyfuel combustion as done in [27]. However, the subsequent complementary determination of flow fields using Particle Image Velocitmetry (PIV) and Laser Doppler Anemometry (LDA), [24] of one air and two oxyfuel lignite flames has opened a new question about the source of differences that have been observed for flow field and particle distribution despite the similar inflow conditions. Thus, for the present configuration the question arises if there is a more general mechanism how momentum changes influence velocity field and flame shape that also holds for different fuels? Appending to the first question a second question arises: if a change in momentum influences flame shape, what is the effect on radiative heat transfer characteristics of the flames? Therefore the present study investigates not only the influence of oxidizer and momentum on flame shape but also varies the employed fuel, hence in addition to the already comprehensively investigated lignite flames, two bituminous coal flames are examined in detail. To draw a comprehensive conclusion of the changes in initial conditions on radiative heat transfer, measurements employing a radiative heat transfer probe have been conducted. Therefore the present paper is not only dedicated to give a detailed image of the studied flames but also to derive a better understanding of the influence of tuning parameters (inflow conditions, fuel and oxidizer) on quantities interesting for retrofitting, such as matching heat flux profiles.