Title: Fluidized Bed Combustion of Manure: Technology Improvement and Sustainability Assessment
Other Titles: Wervelbedverbranding van mest: technologische ontwikkelingen en duurzaamheidsevaluatie
Authors: Billen, Pieter
Issue Date: 17-Feb-2015
Abstract: Because of the shift in agriculture from extensive towards intensive livestock production, in many regions the supply of livestock manure exceeds the demand for nutrients (mainly nitrogen and phosphorus, and to a lesser extent potassium) by crops. Land spreading on agricultural soil is to date still common practice for livestock manure, but may sometimes lead to excessive fertilization of agricultural land, causing environmental problems, e.g. eutrophication. Alternative treatment options are thus needed for large quantities of manure in the EU-28.Combustion of poultry litter, which is a combination of poultry excrements, feathers and bedding material of the stables, in large-scale facilities is feasible because this biomass is relatively dry (less than 45 % of moisture), resulting in a lower heating value of approximately 7 MJ/kg. After combustion, typically in a fluidized bed combustor, recovery of the thermal energy of the hot flue gas in a steam boiler allows the production of renewable, biogenic electricity.The carbon from the poultry litter is biogenic, and therefore the emitted CO2 has no net environmental impact. Moreover, production of electricity avoids the emissions that would have occurred when electricity was generated by fossil fuel combustion. Taking into account the production of auxiliary materials (e.g. lime for flue gas cleaning) and combustion of auxiliary fuels in the poultry litter combustor, the electricity from 1 Mg of poultry litter avoids greenhouse gas emissions corresponding to almost 600 kg of CO2 equivalents, when it replaces electricity from coal combustion. Assuming that the electricity produced from poultry litter combustion replaces electricity from natural gas instead of coal, still more than 240 kg of CO2 equivalents of greenhouse gas emissions are avoided. In contrast, land spreading, the traditional treatment method for manure, does not recover energy from the manure, and therefore the greenhouse gas emissions from the poultry litter spread on land, mainly N2O, result in an increase of the overall greenhouse gas emissions with between 30 kg and 450 kg of CO2 equivalents per Mg, depending on several factors, such as soil type, season and weather. The production of renewable, biogenic electricity in the case of poultry litter combustion, results also in a lower impact in the categories terrestrial acidification, particulate matter formation, eutrophication and fossil depletion, than in the case of land spreading.The combustion ash, containing all of the phosphorus and potassium from the initial poultry litter, can be recycled as an inorganic soil conditioner (this ash is also called a slow release PK-fertilizer). The nitrogen, mainly present as NH3 and urea in the poultry litter, is volatilized during the combustion process, and mainly converted to harmless N2 gas, which is emitted to the atmosphere. To compensate the loss of this nutrient, mineral N-fertilizer such as NH4NO3 with an equal amount of nitrogen as the initial poultry litter could be added to the ash. The production of this NH4NO3 causes the emission of 188 kg of CO2 equivalents of greenhouse gases per Mg of poultry litter combusted, which is lower than the savings from the electricity production. So, even in this scenario, combustion of poultry litter overall avoids emissions of greenhouse gases, and has therefore a lower environmental impact in the category climate change than land spreading.However, the high phosphorus and potassium concentrations of the ash, which are beneficial in respect to ash recycling as a soil conditioner, may be detrimental to the operation of a fluidized bed combustor, typically used for biomass combustion. Previous research showed that for most types of biomass, having a higher potassium concentration (5-15 wt% of the ash) than for instance coal (0.5-3 wt% of the ash), the formation of low melting potassium silicates could result in agglomeration of bed ash, potentially leading to loss of fluidization (defluidization) and subsequent shutdown of the installation. Bed ash consists of the initially added silica sand particles that are coated with ash.Previous reports in literature also showed that, apart from low melting potassium silicates, also potassium and calcium phosphates with a low melting point may be formed in combustion ash and cause agglomeration. However, both silicates and phosphates of potassium and calcium were experimentally observed in combustion ash formed in specific conditions, but it was not clear which salt is formed in which conditions, and which melt dominates in biomass ash. Moreover, thermodynamics did not allow to predict the presence of ternary silicate salts (e.g. CaO-K2O-SiO2 salts) and many phosphate salts, as most databases are incomplete. In this thesis, an appropriate set of thermodynamic data was selected to predict the presence of binary salts of the four major ash elements (Ca, K, P, Si). Thermodynamic data on ternary silicate and phosphate salts and on liquid phases were not used, due to their incompleteness, but these salts and liquids were predicted by using existing phase diagrams. In these phase diagrams (of CaO-K2O-SiO2 and CaO-K2O-P2O5) the formed binary salts were plotted (instead of the total element composition, as was done in previous reports), together with a potential excess of unreacted SiO2, which is abundant due to the use of silica sand as bed material. This means that, as an example, only the fraction of calcium that reacted to calcium phosphate is plotted in the CaO-K2O-P2O5 phase diagram, and only the fraction of calcium that reacted to calcium silicate is plotted in the CaO-K2O-SiO2 phase diagram. This simplified method showed a good agreement with experimental observations and is less susceptible to inaccurate thermodynamic data of complex ternarysalts.It was found that, during fluidized bed combustion of biomass with a high phosphorus concentration, using silica sand as bed material, potassium silicates are preferentially formed over phosphates. These potassium silicates may mix with calcium silicates, and form ternary silicates. Silicate mixtures with a low calcium concentration, and thus rich in potassium, may have very low melting points (e.g. lower than 740 C, which is as low as the typical bed temperature of about 700 C to 750 C), and hence cause agglomeration. However, calcium silicates are thermodynamically less stable than calcium phosphates, with Ca3(PO4)2 being the most stable form. A high phosphorus concentration in the ash therefore enhances agglomeration by lowering the amount of high melting calcium silicates through formation of Ca3(PO4)2. These findings were confirmed by lab agglomeration experiments, in which the concentrations of the four main elements were increased separately in ash samples, after which the samples were subjected to a heat treatment in a muffle furnace. Agglomeration occurred due to the high potassium silicate concentration in the bed ash coatings. This mechanism is called coating induced agglomeration.During the lab agglomeration experiments, a second agglomeration mechanism was discovered, which is called melt induced agglomeration. When H2PO4- or HPO42- salts, which are also present in poultry litter, were added to the bed ash samples, to increase the phosphorus concentration, these salts firstly melted, and then decomposed and reacted with calcium compounds from the ash coatings to form solid Ca3(PO4)2. The molten reactant (H2PO4- or HPO42- salt) is responsible for agglomeration of the bed ash, but the formation of Ca3(PO4)2 causes the initially liquid interparticle bridge to solidify.From the experimental and thermodynamic findings, it appeared that the addition of a calcium salt reduces agglomeration problems. This was also shown in earlier reports, but in this thesis it was demonstrated that the added calcium compound firstly reacts with phosphorus salts in the ash to form Ca3(PO4)2, and only then the excess can effectively react with silica to form high melting calcium silicates, increasing the melting point of the CaO-K2O-SiO2 mixture. Based on these findings, an agglomeration index was developed, to quantitatively determine the amount of calcium salts to be added to avoid agglomeration.The proposed countermeasure, addition of a calcium salt, was applied to a full-scale poultry litter fired fluidized bed combustor. During the test, the silica sand bed material was partially replaced by CaCO3, which decomposes in the fluidized bed to CO2 and reactive CaO. Based on continuous measurements of the pressure drop over the fluidized bed, combined with the value of the agglomeration index, it was shown that the reactive CaO counteracted both agglomeration mechanisms. Coating induced agglomeration was counteracted by the formation of calcium silicates, increasing the melting point of the CaO-K2O-SiO2 mixture in the bed ash. Melt induced agglomeration was counteracted by the presence of small CaO particles, which can react with liquid H2PO4- or HPO42- salts to form solid Ca3(PO4)2, thus avoiding interactions of these liquid salts with bed ash particles. This thesis shows that by applying this countermeasure, sustainable electricity can beproduced from manure by fluidized bed combustion, without severe technological problems.
ISBN: 978-94-6018-967-8
Publication status: published
KU Leuven publication type: TH
Appears in Collections:Process Engineering for Sustainable Systems Section
Materials Technology TC, Campus Group T Leuven
Technologiecluster Materialentechnologie

Files in This Item:
File Status SizeFormat
Doctoraat Pieter Billen.pdf Published 3856KbAdobe PDFView/Open Request a copy

These files are only available to some KU Leuven Association staff members


All items in Lirias are protected by copyright, with all rights reserved.