Evaluating the Combustion Reactivity of Drop Tube Furnace and Thermogravimetric Analysis Coal Chars with a Selection of Metal Additives

Abstract

Opportunities exist for effective coal combustion additives that can reduce the carbon content of pulverized fuel ash (PFA) to below 6%, thereby making it saleable for filler/building material applications without the need for postcombustion treatment. However, with only limited combustion data currently available for the multitude of potential additives, catalytic performance under pulverized fuel (PF) boiler conditions has received relatively little attention. For the first time, this paper therefore compares the reactivity of catalyzed bituminous coal chars from thermogravimetric analysis (TGA) with those generated by devolatilization in a drop tube furnace (DTF). The principal aim was to explore the fundamental chemistry behind the chosen additives' relative reactivities. Accordingly, all eight of the investigated additives increased the TGA burnout rate of the TGA and DTF chars, with most of the catalysts demonstrating consistent reactivity levels across chars from both devolatilization methods. Copper(I) chloride, silver chloride, and copper nitrate were thus identified as the most successful additives tested, but it proved difficult to establish a definitive reactivity ranking. This was largely due to the use of physical mixtures for catalyst dispersion, the relatively narrow selection of additives examined, and the inherent variability of the DTF chars. Nevertheless, one crucial exception to normal additive behavior was discovered, with copper(I) chloride perceptibly deactivating during devolatilization in the DTF, even though it remained the most effective catalyst tested. As a prolonged burnout at over 1000 °C was required to replicate this deactivation effect on the TGA, the phenomenon could not be detected by typical testing procedures. Subsequently, a comprehensive TGA study showed no obvious relationship between the catalyst-induced reductions in the reaction's apparent activation energy and the samples' recorded burnout rates. Although copper(I) chloride did generate a diffusion controlled reaction regime at a lower temperature than the other additives. Furthermore, only the thermally labile iron(III) chloride appeared capable of exerting a catalytic effect under mass transfer controlled combustion regimes, signifying that the physical state of the catalyst could be an important factor during PF combustion.

TGA and Drop Tube Furnace Investigation of Alkali and Alkaline Earth Metal Compounds as Coal Combustion Additives

“…The first of these investigations looked into the combustion acceleration associated with significant quantities of potassium carbonate, 18 while the other compared chars catalyzed by different Group I and II salts with those produced by TGA. 8 Similar investigations have previously contrasted the DTF firing of uncatalysed coal chars with analogous TGA data, but different reactivity trends have often been reported with the results from TGA arguably being closer to those observed in a full-scale boiler, 19−22 even though DTFs are more expensive and time-consuming to run. Conducting burnout studies on a DTF nevertheless appears to be an industry requirement for further examining potential additives, as confirmation of their behavior under a more realistic environment is clearly needed before trials in a pilot plant or boiler can be undertaken.…”

“…26 There was one caveat, however, since the result variability associated with even a moderately effective physical mixture additive has led to a coefficient of variation of between ±33 and ±50% being reported for TGA coal combustion. 8 Therefore, although Figure 2 indicates that all of the tested sodium compounds decreased the time it took for 90% of the char to be converted, none of these results can be described as statistically significant. This data nevertheless allowed the sodium salts to be ranked according to their apparent catalytic performances under these conditions, and on a 1 part additive to 99 parts coal weighting, it produced a decreasing reactivity trend of carbonate > chloride > bicarbonate > sulfate > acetate > nitrate.…”

“…Repeatability and reproducibility were therefore poorer than with the TGA methodology, which demonstrated a coefficient of variation of ±1−3% without a catalyst but an error of ±33−50% with one. 8,26 Previous work has indicated that this unpredictably mainly originates from fluctuating process variables during DTF devolatilization. Likely culprits include the gases' flow rates and pressures, the recycle flow rate as applied by a vacuum pump, the coal feed rate, and any automatic adjustments to the furnace's temperature.…”

“…14−16 These include barium carbonate, cesium chloride, calcium nitrate, and various other alkali carbonates, hydroxides, and nitrates. [8][9][10][11][12]17 However, it currently remains difficult to clarify the underlying chemistry behind these catalysts' performances since, even though many studies have been conducted, very few of them have thoroughly investigated a comprehensive number of compounds in a systematic manner. Researchers have instead chosen to explore catalytic mixtures (such as combinations of metal and semimetal oxides 17 ) or a narrow selection of additives (e.g., just the compounds barium carbonate, iron(III) oxide, and manganese oxide 10 ).…”

Study on Catalytic Combustion of Biomass Mixtures with Poor Coals

“…Alkali, alkaline earth, and transition metals in the form of salts or oxides have demonstrated higher combustion performance by decreasing ignition temperature and increasing combustion rates (Gong et al, 2010;Le Manquais et al, 2011;Wu et al, 1998). The catalytic effect is considered to arise from the improved oxygen transfer behavior, which is caused by the oxygen storage and redox properties of the metal oxides and the first ionization energy of the metal in the case of salts, during the combustion process (Farrow et al, 2013;Gong et al, 2010;Wu et al, 1998).…”