(142c) Characterization of Coal Flame Impacts Under Oxy-Combustion

The future use of coal in the United States depends on technologies being made available to capture and store CO2 emissions from power plants. A key candidate CO2 capture technology is oxy-firing of coal. Using oxygen instead of air to combust the coal produces a flue gas stream that it composed primarily of CO2 and water, making subsequent CO2 capture relatively easy. However, application of oxy-firing to existing power plants presents unquantified challenges as the characteristics of oxy-firing compared to air-firing have not been fully determined.

This presentation will draw results from an on-going DOE-sponsored program on oxy-combustion impacts in existing coal-fired boilers. The primary objective of the program is to develop tools to characterize and predict impacts of CO2 flue gas recycle and burner feed design on flame characteristics (burnout, NOx, SOx and fine particle emissions, heat transfer), slagging, fouling and corrosion, inherent in the retrofit of existing coal-fired boilers for oxy-coal combustion. Key elements to be characterized include:

? Appropriate coal feed, oxygen feed, and flue gas recycle (FGR) design for oxy-fired burners or systems;

? Impacts of oxy-firing system design on boiler flame characteristics (burnout, emissions, heat transfer), fouling, slagging and steam tube corrosion relative to the expected behavior in existing air-fired systems.

The proposed research will accomplish this using:

? New experimental data from Sandia National Laboratories' 0.1 kW Entrained Flow Reactor, University of Utah's 100 kW OxyFuel Combustor (OFC), and University of Utah's 1.5 MW L1500 coal-fired furnace;

? Validated mechanisms describing oxy-firing processes;

? Simulation of oxy-combustion in a pilot-scale furnace and a full-scale utility boiler retrofit using oxy-firing design principles.

Information presented here will include background on oxy-combustion flame mechanisms and results from numerical simulations and experiments conducted in a pilot-scale coal-fired furnace that assess flame behavior under air-fired and oxygen-fired conditions.

Three major issues confronted when considering retrofit of existing boilers for oxy-firing are:

? Selection of the flue gas recycle ratio R, here defined as the ratio of the recycle volume to that of the oxygen. Depending on the retrofit objectives, this could be selected to match heat transfer to furnace surfaces rather than the adiabatic flame temperature, of the air-fired furnace.

? Selection of the location from which to recycle the flue gas, that is, selection of the composition of the recycled flue gas based on the location in the flue gas train from which it is recycled back to the furnace.

? Location of oxygen addition and flue gas recycle streams to the boiler.

The selection of R is determined mainly by interest in maintaining the heat flux rate and distribution as close as possible to those achieved with air combustion in order to minimize any design changes to the boiler. Its value depends on the coal, recycle temperature, whether water is condensed or not. It is generally in the range of 2.7 to 3.3 [Kimura et al, 1995; Payne et al., 1989; Wall, 2007; Croiset et al., 2000]. Differences in air and oxy-firing are in large part determined by the composition of the recycled flue gas, which does not depend on R, but does depend on where the recycle stream comes from. Issues related to retrofit of existing boilers to oxy-coal are driven by these differences, which are significant.

Oxy-coal firing provides additional degrees of freedom in the composition of the transport and secondary oxidant streams, namely PO2 in each stream. In order to obtain maximum advantage of the recycle stream on moderating the combustion temperature the recycle stream should be mixed with oxygen as early as possible. Injection early in the furnace is needed to maximize the heat transfer. It is, however, advantage to use some of the CO2 for coal entrainment into the burner and some of the oxygen for flame stabilization. Testing and simulation can quantify the trade-offs for different injection strategies.

The L1500 pilot-scale combustor at the University of Utah is being used to perform experiments detailing firing system design, flame characteristics, together with selected fouling, slagging and corrosion tests. This 5 MBtu/hr (1.5 MW), pulverized coal-fired furnace produces a combustion environment relevant to full-scale utility boilers. In contrast to lab-scale burners, the turbulent length scales in the L1500 are similar to those associated with commercial burners. The furnace is refractory lined, providing realistic boiler temperatures and radiative environment, necessary for heat transfer and corrosion, despite its 10-fold greater surface-to-volume ratio relative to commercial water wall boilers. The furnace has been retrofitted for oxy-combustion with a flue gas recycle system. This modified facility provides maximum flexibility in where and how the FGR, oxygen and coal may be introduced, including staged conditions for reburning of the recycled NOx.

An oxy-research burner has been designed for use in the L1500 pilot-scale furnace. The initial design of the oxy-research burner was based on a patented commercial-scale burner design. This burner design was reduced to a 1.03 MW (3.5 MBtu/hr) firing rate by using a constant velocity scaling technique. The initial design was evaluated and compared to operation of the existing L1500 low-NOx burner through CFD modeling. The modeling results suggested that the constant velocity scaling method did not produce the desired flame behavior for the burner. To correct this behavior, primary gas velocity was reduced, the quarl was lengthened and the burner face setback relative to the quarl was lengthened. These modifications produced the desired flame stabilization location.

The results of this modeling effort suggest that the dominating parameter governing flame shape and stability is not solely velocity. The flame behavior is closely related to particle heating rate, i.e., in order to maintain similar burner behavior when scaling a burner, the particle temperature profile should be maintained. The particle temperature profile is influenced by many factors including: gas velocity, radiation intensity, particle size and hot flue gas recirculation characteristics inside of the quarl. CFD simulations provided an excellent approach to evaluate the combined impacts of all of these factors.

This work will review the burner design and present results from testing of the oxy-research burner in the L1500 furnace under air-fired and oxygen-fired conditions. Results will focus on flame properties (shape, temperature, composition) and heat flux for different recycle gas compositions and oxygen injection strategies.

References

Croiset, E, Thambimuthu KV, Palmer A. Coal Combustion in O2/CO2 Mixtures Compared with Air. The Canadian Journal of Chemical Engineering 2000; 78: 402 - 7.

Kimura, K, Omata K, Kiga T, Takano S, Shikisima S. Characteristics of pulverized coal combustion in O2/CO2 mixtures for CO2 recovery. Energy Conversion and Management 1995; 36: 805-8.

Payne, R, Chen SL, Wolsky AM, Richter WF. CO2 recovery via coal combustion in mixtures of oxygen and recycled flue gas. Combustion Science and Technology 1989; 67: 1 - 16.

Wall, T. F., Combustion Processes for Carbon Capture, Proceedings Combust. Institute, 31 (2007) 31-47.