Erica LindstrÃm Effect of kaoline

Category: Education

Presentation Description

No description available.


Presentation Transcript


Effect of Kaolin and Limestone Addition on Slag Formation during Combustion of Woody Biomass Pellets Erica Lindström, Dan Boström, Marcus Öhman Energy Technology and Thermal Process Chemistry, Umeå University INTRODUCTION In Scandinavia, the raw materials presently used in pellet production are in short supply, this has lead to an increasing interest in bark and cleaning assortments (branches and tree tops) from the forest industry. High concentration of Si, K and P in the raw material may often led to severe slagging. Therefore it is crucial, in the harvesting and handling of bio energy feedstock, that contamination from sand and dirt (which increases the Si-levels substantially) is avoided. Several suggestions of mineral additives, that could prevent slagging have been made during the last 15 years [1], [2], [3] and [4]. The most promising additives are kaolin and limestone. OBJECTIVES The objectives in the present work are: (i) To determine the effects of kaolin and limestone addition upon slagging tendencies of problematic woody biomass pellets during combustion in a small scale residential pellet burner (20 kW), (ii) to contribute to the understanding of the role of kaolin and limestone in preventing slagging. EXPERIMENTAL FUELS AND ADDITIVES Two severely contaminated raw materials, bark and cleaning assortment, were used. The additives used were limestone slurry (78 wt %d.s.) and kaolin slurry (66 wt %d.s.), with a particle size of 1-2 µm. Initial combustion experiments showed that a moderate addition of 1 wt %d.s. of kaolin and 2 wt %d.s. of limestone was sufficient to reduce the slag formation. Element analyses for the six pellet assortments are given in Table I. COMBUSTION The combustion study was performed in a commercial under fed pellet burner (20 kW) installed in a reference boiler used for the national certification test of residential pellet burners in Sweden. The test runs lasted for 19 ±2 hours corresponding to a total burned pellet amount of 34.9 ± 8.9 kg. Combustion temperatures were measured continuously on and in the vicinity to the burner grate. Continuous measurements of O2, CO, NO and SO2 were also performed with conventional instruments in the exhaust gas directly after the boiler. The amounts of deposited ash and slag in the burner as well as in the boiler were quantified after every experiment and the products collected for analysis. CHEMICAL ANALYSES The collected deposits from the experiments were characterized with X-ray diffraction (XRD), scanning electron microscopy (SEM) combined with energy –dispersive X-ray analysis (EDS). All the deposits from the boiler and the burner were sieved to separate ash from slag. All melted particles greater than 0.5 cm was removed from the ash as slag. Energy Technology and Thermal Process Chemistry Umeå University SE-901 87 Umeå, Sweden Phone: +46 (0)90-786 59 71 Fax: +46 (0)90-786 91 95 E-mail: Financial support from STEM is gratefully acknowledged RESULTS AND DISCUSSION COMBUSTION CONDITIONS The average combustion temperature was estimated to be 1133 ±93 oC. The CO emissions were 2265 ±778 ppm. The Oxygen level was 11 ±2%. Considering all the known combustion data i.e. temperature, time, fuel amount, CO emissions and O2 level, the assumption that the combustion conditions were similar during all experiments was made. DEPOSITED SLAG A comparison between the three different bark pellet assortments showed a reduction of slag formation with 99 and 92% with kaolin and limestone respectively. An equivalent comparison between the three cleaning assortment pellets shows a reduction of slag formation with 30 and 100% with kaolin and limestone respectively. Figure I shows the amounts of slag produced from the different assortments. All the slag showed approximately the same degree of sintering i.e. hardness; mostly partly melted ash that still could be broken up in pieces with elements of totally melted ash i.e. glass. Table I: Fuel characteristics and main ash forming elements of the studied fuels. Figure I: Fraction of in going fuel ash that formed slag. CONCLUSIONS Adding limestone suspension with additive-to-fuel ratio of 1 wt %d.s. to the severely contaminated bark and cleaning assortment clearly lowered the slagging tendencies. Adding kaolin suspension with additive-to-fuel ratio of 2 wt %d.s. to the severely contaminated raw materials, the slagging tendencies were also clearly lowered. Over all, the results from limestone addition were more satisfying regarding the slag formation than kaolin addition. REFERENCES [1] Öhman, M.; Boström, D.; Nordin, A.; Hedman, H., Effect of kaolin and limestone addition on slag formation during combustion of wood fuels. Energy & Fuels 2004, 18, (5), 1370-1376. [2] Steenari, B. M.; Lindqvist, O. High-temperature reactions of straw ash and the anti-sintering additives kaolin and dolomite. Biomass and Bioenergy 1998, 14, (1), 67-76. [3] Turn, S. Q., et al. A review of sorbent materials for fixed bed alkali getter systems in biomass gasifier combined cycle power generation applications. J.Inst. Energy 1998, 71, 163-177. [4] Öhman, M.; Nordin, A. The role of kaolin in prevention of bed agglomeration during fluidized bed combustion of biomass fuels. Energy & Fuels 2000, 14, (3), 618-624. ****, dominant (>50 % of sample); ***, subdominant (20-50 % of sample); **, minor (5-20 % of sample); *, trace (<5 % of sample). All units are in mg/kg d.w. if not otherwise indicated. * %. ** % d.w. Figure III: Element distribution of the formed slag presented on carbon and oxygen free basis. Table II: Phases identified in the bottom ash and slag. CHEMICAL COMPOSITION The collected samples of bottom ash and slag were analyzed with SEM/EDS to determine the main elements in the respective deposits. Figure II shows a typical image of a polished cross section of a slag sample in epoxy resin. It shows that the melted bottom ash “glues” the sand particles together. The epoxy resin is between the slag particles and in the cavities of the slag. The average main elements in the ashes are Si, Ca, Al and K, given in descending order. Average compositions of the slags are given in Figure III. Areas of 200*200 µm of the slag were analyzed. Only areas of apparent melted bottom ash were analyzed (Figure II). Dominating elements in both the bottom ash and the slag were Si, Ca, Al and K. The collected samples of bottom ash and slag were further analyzed with XRD for identification of crystalline phases. The presence and approximate amount of detected phases in the slags are given in table II. In the ashes the dominant crystalline phase found were quartz, minor phases found were microcline and albite and traces of åkermanit, leucite wollastonite, hematite (Fe2O3) and calcite (CaSiO3) were also found. Figure II: A typical image of a polished cross section of a slag sample in epoxy resin.

authorStream Live Help