Demonstrating Fuel Design To Reduce Particulate Emissions and Control Slagging in Industrial-Scale Grate Combustion of Woody Biomass

Abstract

The demand for increased overall efficiency, improved fuel flexibility, and more stringent environmental legislations promotes the development of new fuel- and technology-related concepts for the bioenergy sector. Previous research has shown that careful consideration of the fuel ash composition and the adjustment of the same via various routes, i.e., fuel design, have the potential to alter the ash transformation reactions, leading to, e.g., a reduction of the formation of slag or entrained inorganic ash particles. The objective of the present work was, therefore, to demonstrate the use of fuel design as a primary measure to reduce the emission of PM1 during combustion of woody biomass in medium-scale grate-fired boilers while keeping the slag formation at a manageable level. This was achieved by designing fuel blends of woody biomass with carefully selected Scandinavian peats rich in Si, Ca, and S. The work includes results from three experimental campaigns, performed in three separate grate-fired boilers of different sizes, specifically 0.2 MWth, 2 MWth, and 4 MWth. In one of the campaigns, softwood-based stemwood pellets were copelletized with different additions of peat (5 and 15 wt %) before combustion. In the other campaigns, peat was added in a separate fuel feed to Salix chips (15 wt % peat) and softwood-based stemwood pellets (10 and 20 wt % peat). Particulate matter and bottom ashes were characterized by scanning electron microscopy–energy-dispersive X-ray spectroscopy for morphology and elemental composition as well as by powder X-ray diffraction for crystalline phase composition. The results show that the fuel design approach provided PM1 reduction for all fuel blends between 30 and 50%. The PM1 reduction could be achieved without causing operational problems due to slagging for any of the three commercial boilers used, although an expected increased slagging tendency was observed. Overall, this paper illustrates that fuel design can be implemented on an industrial scale by achieving the desired ash transformation reactions, in this case, leading to a reduction of fine particulate emissions by up to 50% without any operational disturbances due to slag formation on the grate.

Will mixing rule or chemical reactions dominate the ash behavior of biomass mixtures in combustion processes on laboratory and pilot scales?

“…However, the main focus in this work was on the reaction within the ash of the fuel mixtures. The mechanism of reactions are so complex and highly depend on the ash composition and the operating conditions [5,6,28]. Nevertheless, since the ash used in this work were mostly composed of Si, Ca, and K (Table 1), chemical reactions, if taking place, will be governed by the interactions between these three elements in the form of oxides.…”

“…Depending on the ash composition of the fuel and in the case of high availability of K in the form of KOH, the density of molten potassium-silicates particles increases significantly. But as the fuel particles are burning out, it shrinks and the ash particles approach each other with an increasing opportunity for contact [6,28]. This will lead to an aggregation of the molten potassium-silicate particles, which will form a larger droplets and initiate slag formation.…”

“…The chemical reactions within the ash of the biomass mixture, which highly depend on operating conditions, were the interest of several researchers in the field [6,21,[25][26][27], especially in laboratory-scale experiments. Some of them suggested various reaction pathways that might take place between the inorganic elements of the ash of biomass [6,28,29].…”

“…Several researchers proposed the use of a thermodynamic tool, i.e. the CaO-K2O-SiO2 ternary phase diagram to predict ash behavior, such as Rebbling et al [28], Zeng et al [21], Defoort et al [27], and Ohman et al [7]. Despite the relative success of such a prediction tool, the same authors concluded that the thermodynamic database needs to be revised since some crystalline phases appeared experimentally but were absent in the database [27], and the calculated liquidus line differed from the experimental results [21,28].…”

“…the CaO-K2O-SiO2 ternary phase diagram to predict ash behavior, such as Rebbling et al [28], Zeng et al [21], Defoort et al [27], and Ohman et al [7]. Despite the relative success of such a prediction tool, the same authors concluded that the thermodynamic database needs to be revised since some crystalline phases appeared experimentally but were absent in the database [27], and the calculated liquidus line differed from the experimental results [21,28]. The main drawback of the thermodynamic prediction tool is that it provides results at equilibrium, which might or might not be the case in real processes, especially on an industrial scale.…”