The influence of particle size distribution on bio-coal gasification rate as related to packed beds of particles

Sammanfattning: This thesis is a part of a collaboration between Höganäs AB and Luleå University of Technology, aiming at replacing fossil process coal with bio-coal in their sponge iron process. The difference in gasification reactivity, i.e. reaction rate, between fossil coals and bio-coals is the major challenge in the endeavor to decrease the climate impact of the existing process. The goal of this thesis is to develop a model of reaction rate for bio-coals in relation to particle size distribution. Different particle size distributions were combined and tested to see how that affects the effective reaction rate. Within the scope of this work, gasification reactivities of different materials, including coal, cokes, and bio-coals, were determined. Three bio-coals were selected to study the effect of particle size distribution on reactivity. Kinetic parameters were determined by using thermogravimetric analysis in the temperature range of 770-850 °C while varying CO2 partial pressure between 0.1-0.4 atm. The effect of particle size on the reaction rate was investigated by using particles with diameter between 0.18 and 6.3 mm. The effect of particle size distribution on the reactivity of bio-coal in a packed bed was carried out in a macro thermogravimetric reactor with a constant bed volume of 6.5 cm3 at 980 °C and 40% (vol.) of CO2. The experimental investigation in three different rate-limiting steps was done for one bio-coal sample, i.e. Cortus Bark bio-coal. The activation energy of the bio-coal was 187 kJ mol-1, and the reaction order was 0.365. For the internal diffusion control regime, an increase in particle size resulted in low reaction rate. The effective diffusivity calculated from the Thiele modulus model was 1.41'10-5 m2 s-1. For the external diffusion control regime, an increase in particle size increased the reaction rate up to a certain point where it plateaued at >1 mm. By choosing two discrete particle size distributions, where a smaller average distribution can fit into a larger average distribution the reaction rate was lowered by 30% compared to only using a single narrow particle size distribution. This solution decreased the difference of apparent reaction rate in a packed bed between the bio-coal and anthracite from 6.5 times to 4.5 times. At the moment the model is not generalized for all bio-coals. However, the developed methodology can be routinely applied to assess the different bio-coal samples. One possible error can be that pyrolysis influences the gasification rate for bio-coal that is pyrolyzed below the temperature of the gasification test. There is a clear correlation between particle size distributions, bulk density, and apparent reactivity. By mixing two distributions the reaction rate of Cortus Bark was reduced from 6.5 times the reaction rate of anthracite to 4.5.