The release of sodium during the combustion of black liquor is a significant source of fume formation in a kraft recovery boiler, affecting efficiency in a pulp and paper mill. The fume is deposited on the surface of heat exchanger tubes in the upper furnace, causing fouling and corrosion, especially to the superheaters. This thesis reports on work done to develop improved understanding of fume formation. The mechanisms of sodium release during each stage of black liquor combustion are influenced by the surface temperature. The addition of boron to the black liquor, which debottlenecks the recausticizing plant by a reduction in lime usage, also influences the characteristics of black liquor combustion, such as combustion time and swelling. Previously, no effective measurement technique has been available to quantify sodium concentration in the plume of a burning black liquor droplet with or without boron, or to record the distribution of surface temperature through the time history of a burning droplet. This thesis reports on the adaptation of two techniques for the measurement of the release of atomic sodium and the temperature history, and their application to investigate several aspects of the release of atomic sodium during combustion of black liquor in a flat flame environment. The simultaneous employment of a planar laser-induced fluorescence (PLIF) technique with an absorption technique has been adapted to allow quantitative measurement of the release of atomic sodium. The absorption technique has been employed to correct for both fluorescence trapping due to absorption and attenuation by high concentration of the atomic sodium in the plume, and for collisional quenching by the other major gas components present in the flat flame. An independent assessment was performed using kinetic calculations, based on measured total sodium that is residual in a particle obtained at different stages in the combustion process. These independent assessments were used to provide greater insight in to the release process and to cross-check. The influence of both the initial diameter of the droplet and addition of boron to the black liquor on the temporal release and the release rate of atomic sodium during the combustion have been performed using the present PLIF technique. The second technique, two-dimensional two-colour optical pyrometry, has been adapted to measure the distribution of surface temperature and the swelling (change in surface area) of a burning black liquor droplet. The influence of surface temperature or the change in the external surface area of the droplet on the release of atomic sodium during the combustion of black liquor has been assessed through concurrent use of both adapted techniques. The highest concentration of atomic sodium was measured in the final stage of combustion that of smelt coalescence, where it is an order of magnitude greater than in the other stages combined. While the extensive release of atomic sodium at high temperature in this final combustion stage occurs in only a relatively small percentage of droplets in a kraft recovery boiler, the effect could still be significant in fume formation. This is because the extensive release is expected to occur in the very small droplets, predominantly generated by splitting or physical ejection. Small droplets will have a very short combustion time and so could remain in suspension within hot gases for sufficient time for extensive release of sodium. These measurements outcomes can be used to support the future development of sub-models for computational fluid dynamics (CFD) models in order to better understand and optimise fume formation in a kraft recovery boiler. / http://proxy.library.adelaide.edu.au/login?url= http://library.adelaide.edu.au/cgi-bin/Pwebrecon.cgi?BBID=1474431 / Thesis (Ph.D.) - University of Adelaide, School of Mechanical Engineering, 2009
Identifer | oai:union.ndltd.org:ADTP/288238 |
Date | January 2009 |
Creators | Saw, Woei Lean |
Source Sets | Australiasian Digital Theses Program |
Detected Language | English |
Page generated in 0.0016 seconds