Formation and removal of biomass-derived contaminants in fluidized-bed gasification processes:Licentiate thesis

The objectives of this thesis were to examine the effects of the feedstock and the operating conditions of a fluidized-bed gasifier on the formation of tars and nitrogen-containing compounds and to study the effectiveness of the hot gas cleaning methods developed for the removal of particulates, alkali metals, tars and nitrogen-containing compounds. The most essential part of the work was carried out in the pressurized fluidized-bed gasification test facilities composed of an air-blown bubbling fluidized-bed gasifier and subsequent hot gas filter unit. The operation pressure of the test rig could be varied in the range 0.3 - 1.0 MPa and the maximum allowable gasification temperature was 1 050 oC. The maximum capacity with biomass fuels was 80 kg/h. A wide range of feedstocks from hard coals, lignite and peat to different wood-derived fuels and straw were used in the gasification tests. Very little tars were formed in the gasification of hard coals and lignites. In peat gasification 1 - 3 % of the dry ash free matter of the feedstock was converted into tars. In wood gasification the tar yields were roughly five times higher than in peat gasification. With all feedstocks studied, the total tar content could be clearly reduced by increasing the gasification temperature. At low temperatures (below 800 oC) the tar consisted of a wide range of different organic compounds, while at above 850 oC all unstable pyrolysates were decomposed leaving only benzene and higher-molecular-mass aromatic compounds. The formation of problematic high-molecular-mass compounds (>200 g/mol) was typical of gasification of biomass fuels with fine particle size distribution (such as saw dust). In the gasification of wood chips and other biomass fuels with a larger particle size, the formation of high-molecular-mass tars was avoided by using dolomite as the bed material. The fate of fuel nitrogen in fluidized-bed gasification tests was dependent, on one hand, on the gasification temperature and air ratio and, on the other hand, on the feedstock properties. With all feedstocks studied ammonia was the main fixed nitrogen species present in the product gas. With high-volatile fuels, generally gasified at low temperatures, more than 60 % of feedstock nitrogen was converted into ammonia. Clearly lower conversions were measured for high-temperature gasification of bituminous coals. Two different types of ceramic filters were tested in the filter unit connected to the pressurized fluidized-bed gasifier. The filter unit was operated in a temperature range of 400 - 740 oC. The particulate removal requirements set by the gas turbines were met by both types of filters and with product gases derived from all the feedstocks tested. However, the early experiments with the homogenous filters (uniform pore size throughout the filter element) resulted in slowly but continuously increasing pressure drop. Constant pressure drop behaviour was later achieved with product gases obtained from coal and peat gasification using two-layer filter elements, which consisted of a thin fiber-containing layer of small pores outside the coarser support structure. The first saw dust gasification tests led to almost complete blocking of the filters in less than five hours. Satisfactory filtration behaviour was later achieved also in saw dust gasification by optimizing both the gasification and filtration conditions and by using dolomite addition into the gasifier. In the fluidized-bed gasification tests the bulk of alkali metals was retained in the solid output streams (bottom ash, particulates removed by the cyclone and the filter unit). Only less than 1 % of the feedstock alkalis was found in the hot product gas leaving the secondary cyclone. However, the vapour-phase concentrations determined after the hot cyclones were generally of the order of 1 ppm-wt, while the estimated allowable maximum alkali concentration in the product gas was of the order of 0.1 ppm-wt. This very stringent limit set by the gas turbine manufactures could, however, be met with most gasification feedstocks by cooling the product gas to below 400 - 550 oC before the ceramic filter unit. In addition to the gasification and gas filtration tests, catalytic tar and ammonia decomposition was studied using both laboratory and bench-scale test facilities. Inexpensive calcium-based bulk materials, dolomites and limestones, were efficient tar decomposition catalysts in atmospheric-pressure tests. However, these materials lost their activity at high operating pressures, where the calcination of CaCO3 was prevented by the high partial pressure of CO2. Calcium-based materials did neither have catalytic effects on ammonia decomposition. Nickel-based materials were effective both at atmospheric-pressure and high-pressure tests. Total tar decomposition and 80 - 90 % reduction in ammonia concentration were obtained by the novel nickel-based monolith catalyst tested (at 0.5 MPa, 900 oC) in a slip stream of the pressurized gasifier. The activity of the catalyst also remained constant throughout the 500-hour extended time test, during which the operating conditions were kept constant