Effect ofInitial Potassium Catalyst Concentrationon Its Activity and Fate in CO₂ Gasification of LigniteChar

K-catalyzed gasification of char and carbon has been studied for more than 40 years while many different types of relationships between specific rate of gasification (<i>r</i>_sp = dXdt(1−X), <i>X</i>; char conversion) and K concentration in gasifying char/carbon (<i>C</i>_K) have been reported. This work explored the mechanism causing such diversity in the <i>r</i>_sp–<i>C</i>_K profile. An ash-free lignite was impregnated with K₂CO₃, devolatilized, and then gasified with CO₂ at 800 °C and <i>C</i>_K at the start of gasification (<i>C</i>_K,0) of 0.15–2.66 mol-K/kg-daf-char. <i>C</i>_K was accurately determined by measuring the K volatilization rate as a function of <i>X</i>. For <i>C</i>_K,0 = 0.15, <i>r</i>_sp increased linearly with <i>C</i>_K while the K catalyst activity (<i>k</i>_c′ = <i>r</i>_sp/<i>C</i>_K) remained unchanged at 0.013 min^–1 mol-K/kg-daf-char^–1. The K species underwent transformation to much more active catalyst (<i>k</i>_c′ ≈ 0.130 min^–1 mol-K/kg-daf-char^–1) for <i>C</i>_K,0 = 0.64–2.66, where the <i>r</i>_sp–<i>C</i>_K profile was independent of <i>C</i>_K,0, and <i>r</i>_sp reached an upper limit due to saturation of gasifying char matrix with the catalyst. The catalyst transformation also occurred for <i>C</i>_K,0 = 0.28–0.39 while <i>k</i>_c′ of the transformed catalyst was steady at 0.03–0.08 depending on <i>C</i>_K,0. Thus, the variety of the <i>r</i>_sp–<i>C</i>_K profile arose from that in <i>C</i>_K,0. Higher <i>C</i>_K,0 led to greater d<i>X</i>/d<i>t</i> and then d<i>C</i>_K/d<i>t</i> under a steady supply of oxygen (as CO₂) that anchored K on the char surface, allowing atomic-level dispersed C–O–K and/or K₂CO₃ to transform into clusters consisting of K_<i>x</i>O_<i>y</i>(CO₂)_<i>z</i> (<i>y</i>/<i>x</i> > 0.5, <i>z</i>/<i>x</i> ≈ 0.16). The most active cluster was formed even during the period of devolatilization for <i>C</i>_K,0 > 1.69. Increasing <i>C</i>_K,0 from 0.15 to 2.66 (by a factor of 18) resulted in shortening of the time required for X = 0.50 and 0.99 by factors as large as 1/200 and 1/100, respectively, due to the above-mentioned <i>C</i>_K,0 effect on the <i>r</i>_sp–<i>C</i>_K relationship.