Slag formation in fixed bed combustion of phosphorus-poor biomass

Abstract: To handle a great demand for biomass, alternative biomasses beyond stem wood are being introduced into the solid fuel combustion market, fuels with generally higher (>0,5 wt-%) ash content and different fuel ash compositions compared to stem wood, such as forest residue, bark, grass and straw. Unfortunately, combustion of these alternative fuels often causes more ash related problems such as fouling, slagging and higher particle emissions compared to combustion of stem wood. Many research studies have been conducted regarding ash melting and ash sintering in biomass combustion. However, literature discussing slagging of biomass ash is rather scarce, especially relating to fixed bed combustion. The majority of the biomass fuels available on the market today are phosphorus-poor and this thesis emanates from those. The overall objective was to obtain knowledge of slag formation in fixed-bed combustion of phosphorus-poor biomass, based on bench- and full-scale experiments, chemical analysis of produced ash fractions, chemical equilibrium calculations, viscosity estimations and statistical evaluations. This thesis investigates slagging of [phosphorus-poor] biomass in fixed bed combustion. 85 fuels and 10 different burner/boiler technologies were utilized. The results in this thesis highlight the importance of the ash forming elements Si, Ca, K and Alin the slag formation process in fixed bed combustion of phosphorus-poor biomass. Increased Ca/Si ratios in the fuel reduce slag formation due to formation of more temperature stable phases, i.e. Ca/Mg-oxides and/or formation of carbonate melts with lower viscosity (not sticky) that are less prone to forming slag. A high Al/Si ratio increases the possibility of forming solid and thermally stable K−Al silicates that can reduce slag formation. The fraction of ash melt, along with viscosity, are critical for slag formation and these parameters vary between different fuels. Four classes were defined according to their slagging potential; 1) No slag: fuel composition and the bottom ash contains low Si and K contents and higher Ca content. Fuel examples: non-contaminated stem- and pulpwood/energy wood, 2) Minor slagging tendency: fuel compositions show increased Si compared to non-slagging fuels and the bottom ash contains lower Ca, but increased Si content and approximately unchanged K content compared to the former category. Fuel examples: stem wood, bark and logging residue with increased Si-content due to light contamination. 3) Moderate slagging tendency: fuel composition contains further increased Si content. Increased share of formed silicate melt and higher viscosity (more sticky) compared to minor slagging fuels. Fuel examples: mostly contaminated woody fuels and grass and straws with relatively high amount of Ca. and 4) Major slagging tendency: Fuel composition contains high Si and K content. Sticky K-silicates causes major increase in slagging tendency. Fuel examples: different types of grass and straw fuels. The burner/boiler technology can affect whether slagging will induce major problems in the burner or not. However, long residence times and high temperatures for the combustion residues in the hot part of the fuel bed are technical prerequisite for increased slag formation. This thesis developed two qualitative fuel indices for predicting slagging in fixed bed combustion of phosphorus-poor biofuels – one index for fraction of fuel ash that forms slag and one index for sintering category of the formed slag. Both novel indices deliver acceptable results and are more reliable than previous indices found in the literature. Importantly, the fraction of fuel ash that forms slag index outperforms the sintering category for qualitative prediction of the problematic slagging potential of a certain fuel. Additional work is needed to further widen the compositional range as well as to fine tune the indices’ boundaries.