Abstract
The increasing interest towards more efficient and clean technologies, specially paying attention to CO2 neutral processes is encouraging the investigation of coal thermochemical conversion under oxy-fuel atmospheres (i.e. without N2). The process offers many advantages such as easy separation of the CO2 produced and low NOX/SOX emissions. While coal conversion in air is already well understood, full understanding of the influences of CO2-rich atmospheres is still required. A series of experiments in thermogravimetric analyser, drop-tube reactor and flat-flame burner were performed using a mid-range bituminous coal (Colombian coal) to understand the differences in the chars obtained after the pyrolysis step. Comparing the pyrolysis in N2 and CO2 atmospheres, significant differences were observed in the resulting chemical (composition) and physical properties of the chars, whereas mass loss was very similar for short residence times (< 130 ms). Afterwards, the chars obtained were submitted to oxidation and gasification, under several different operating conditions, in order to evaluate the difference in reactivity of these chars. Chars obtained under CO2 atmospheres revealed a lower reactivity, despite their higher surface area. These aspects cannot be explained and captured by a model which does not focus on some important details. In this paper, the results obtained in these experiments are summarized and discussed on the point of view of the POLIMI modelling approach for thermochemical conversion of solids. This model offers several advantages, such as being flexible to improvements, requiring simple experimental data of the fuel and offers an all-in-one solution for describing the kinetics of the whole process. The developments accounted for a wide range of experimental data, which allowed its calibration for several fuels, mostly in air combustion. It was first developed to describe the pyrolysis step, and later char oxidation/gasification was included in a simplified approach. The detailed mechanism of homogeneous gas phase reactions of the volatiles is also coupled. In order to extend the predictive capabilities of the model for oxy-fuel conditions, dedicated experiments must be considered for future improvements. In this work these missing effects are discussed, identifying the main necessary improvements, allowing this model to be extended and applied also for the designing of reactors that use oxy-fuel technologies.