This paper is part I in a series of two describing the fate of alkali metals and phosphorus during cocombustion of rapeseed cake pellets in a 12 MW thermal CFB boiler. In paper I the results of using the mixture of wood chips and wood pellets as a base fuel are described. Up to 45% on energy basis of rapeseed cake was cocombusted during a 4 h test. Two approximately 12 h tests with energy fractions of rapeseed cake of 12 and 18% were performed with limestone as a varying parameter. Fuels were characterized by means of chemical fractionation and standard methods. Elemental mass balances were calculated for ingoing and outgoing streams of the boiler. In addition SEM/EDX analyses of ashes were performed. Gaseous (KCl þ NaCl) as well as HCl and SO 2 were measured upstream of the convection pass, where deposit samples were also collected with a deposit probe. The deposit samples were analyzed semiquantitatively by means of SEM/EDX. The elemental mass balances show accumulation of alkali metals and phosphorus in the boiler. Analyses of bed material particle cross sections show the presence of phosphorus compounds within a K-silicates matrix between the agglomerated sand particles, indicating a direct attack of gaseous potassium compounds on the bed surface followed by adhesion of ash particles rich in phosphorus. Build-up of deposit during the cocombustion tests mainly took place on the windward side of the probe; where an increase of K, Na, and P has been observed. Addition of limestone prevented formation of K-silicates and increased retention of phosphorus in the bed, most probably due to formation of high-melting calcium phosphates. During the tests with limestone, an increase of potassium chloride upstream of the convection pass and a decrease of phosphorus in the fly ash fraction could be noticed. Agglomeration and slagging/fouling when cofiring wood with rapeseed cake may be linked to its high content of organically bonded phosphorus;phytic acid salts;together with high contents of water-soluble alkali metals chlorides and sulfates in the fuel mixture.

The bed agglomeration characteristics resulting from the combustion of 11 mixtures of rapeseed cake and spruce bark were studied in a bench-scale bubbling fluidized-bed reactor (5 kW). The objective was to determine the defluidization temperatures and the prevailing bed agglomeration mechanism as functions of the fuel mixture. Controlled fluidized-bed agglomeration tests were performed for each mixture with quartz sand as the bed material. The total defluidization temperatures and the initial defluidization temperatures were determined based on the measured pressure and temperature profiles in the bed. After combustion, bottom ash samples, agglomerates, and fly ash samples were analyzed by means of scanning electron microscope combined with energy dispersive X-ray detector (SEM-EDX). The composition of the ash-forming matter produced by the combustion of rapeseed cake is significantly different from that produced by the combustion of bark, resulting in different bed agglomeration tendencies. Bark contains ash-forming matter dominated by calcium, with some silicon and potassium, whereas rapeseed cake is rich in phosphorus, potassium, and sodium. The total defluidization temperature for pure bark was above 1045 °C, whereas, for rapeseed cake, defluidization occurred during combustion (800 °C). During the combustion of bark, the formation of a potassium-rich layer on the silica-bed grains was found to be a crucial for the formation of agglomerates. The low defluidization temperature for the rapeseed cake can be attributed to the formation of sticky ash, which is dominated by phosphates. Two main phosphate forms were observed in the neck between the silica grains: calcium–potassium/sodium phosphates, and magnesium–potassium phosphates. As the proportion of bark increased, the Ca/P ratio increased in the fuel mixture, and the formation of high-temperature melting phosphates in the ash was favored. However, the addition of bark also favored the formation of a potassium-rich layer on the silica bed material, leading to the coexistence of both bed agglomeration mechanisms. In the present work, mixtures with a minimum of 60 wt % bark resulted in significantly increased defluidization temperatures and reduced bed agglomeration tendencies, compared to what occurs in rapeseed cake monocombustion.

Recently the use of algae for CO 2 abatement, wastewater treatment and energy production has increasingly gained attention worldwide. In order to explore the potential of using algae as an alternative fuel as well as the possible challenges related to the algae gasification process, two species of macroalgae, Derbesia tenuissima and Oedogonium sp., and one type of microalgae, Scenedesmus sp. were studied in this research. In this work, Oedogonium sp. was cultivated with two protocols: producing biomass with both high and low levels of nitrogen content. Cogasification of 10 wt% algae with an Australian brown coal were performed in a fluidized bed reactor and the effects of algae addition on syngas yield, ash composition and bed agglomeration were investigated. It was found that CO and H 2 yield increased and CO 2 yield decreased after adding three types of macroalgae in the coal, with a slight increase of carbon conversion rate, compared to the coal alone experiment. In the case of coal/Scenedesmus sp, the carbon conversion rate decreased with lower CO/CO 2 /H 2 yield as compared to coal alone. Samples of fly ash, bed ash, and bed material agglomerates were analysed using scanning electron microscopy combined with an energy dispersive X-ray detector (SEM-EDX) and X-ray diffraction (XRD). It was observed that both the fly ash and bed ash samples from all coal/macroalgae tests contained more Na and K as compared to the coal test. High Ca and Fe contents were also found in the fly ash and bed ash from the coal/Scenedesmus sp test.Significant differences in the characteristics and compositions of the ash layer on the bed particles were observed from the different tests. Agglomerates were found in the bed material samples after the co-gasification tests of coal/ Oedogonium N+ and coal/ Oedogonium N-. The formation of liquid alkali-silicates on the sand particles was considered to be the main reason of agglomeration for the coal/ Oedogonium N+ and coal/ Oedogonium N-tests. Agglomerates of fused ash and tiny silica sand particles were also found in the coal/ Scenedesmus sp test. In this case, however, the formation of a Fe-Al silicate eutectic mixture was proposed to be the main reason of agglomeration. Debersia was suggested to be a potential alternative fuel which can be co-gasified with brown coal without any significant operating problems under the current experimental conditions. However, for the other algae types, appropriate countermeasures are needed to avoid agglomeration and defluidization in the co-gasification process.

Biomass residues and wastes will have an important role in the future energy mix. Residues from agriculture or industrial processes are usually cheap, widespread, and continuously produced, and their heating value is comparable to that of wood. However, many of these materials are rich in proteins and, thus, nitrogen. Therefore, to be able to exploit these materials in an efficient way, fundamental research on their nitrogen chemistry is essential. In the work presented here, the release of gaseous compounds from five biomass residues from different processes was tested in a bench-scale single-particle reactor. The tested fuels were dry distiller's grains and solubles (DDGS), palm kernel cake (PKC), rapeseed cake (RC), fermented sewage sludge (FSS), and chicken manure (CM). All of these materials have sufficiently high heating values and are either already used or projected to be used soon in industrial-scale boilers. The setup allowed us to study the total C and N release and the nitrogen partitioning between volatiles and char. Release profiles of CO, CO 2 , and NO for each fuel were measured at 800, 900, and 1000°C and 3 and 10 vol % O 2 . DDGS, PKC, and RC showed similar release profiles, while FSS and CM reacted faster because of a lower content of fixed C and faster char oxidation because of the catalytic effect of the (high) inorganic content. The maximum amounts of carbon dioxide released from these fuels, 80-100 g of CO 2 /MJ, are similar to the levels for coals. The value of total N released as NO was minimum for FSS, 0.3 g of N/100 g dry basis (db), equal to around 0.5-0.6 g of N/100 g db for DDGS, PKC, and CM, and maximum for RC, equal to 0.8 g of N/100 g db. The total fuel N to NO conversion showed a trend similar to what was reported in past studies: the fuel with the lowest amount of fuel N, PKC in this study, released the most of initial N as NO, while the fuel with the highest fuel N content, CM, presented the lowest conversion. From the current analysis, it appears that under mild conditions, at 800°C, all of the fuels, except FSS, released carbon and nitrogen proportionally. FSS released most of its nitrogen during devolatilization at all temperatures, while for the other fuels, the amount of volatile N increased significantly with the temperature. The data presented here seem to indicate that the large amount of fuel N contained in the tested fuels could resort in enhanced thermal De-NO x reactions, which, with the correct operational conditions, could keep NO emissions below legal limits even without additional gas-cleaning equipment.

This paper is part 2 in a series of two papers describing the fate of alkali metals and phosphorus during cocombustion of rapeseed cake pellets with different fuels in a 12 MW th CFB boiler. In the first part (