화학공학소재연구정보센터
Fuel Processing Technology, Vol.134, 424-440, 2015
A numerical model for understanding the behaviour of coals in an entrained-flow gasifier
This paper presents the development of a practical and flexible steady-state gasification model in which existing mechanistic models are incorporated with new knowledge of gasification reaction fundamentals. In particular, intrinsic char gasification kinetics and rates of char gasification are reconciled using a 'composite effectiveness factor' by taking into account morphological types of char particles and their impact on char conversion rate. Flows inside an entrained flow gasifier are simplified using ideal chemical reactors consisting of two plug flow and two well stirred reactors. Whilst clearly a simplification of a complex entrained flow gasifier, this approach accounts for the three dimensional aspects of recirculation and mixing and allows rapid convergence as required for incorporation into a practical IGCC process model. Experimental data from gasification of the same coals in a 5 MWth entrained flow gasifier were used to validate the performance of the model. Model calculations of the impact of oxygen-carbon stoichiometry on char conversion, cold gas efficiency (CGE) and product gas composition, using laboratory-scale measurements as inputs, are consistent with measurements at pilot-scale. The model results show that maximum CGEs for the higher reactivity coals with relatively high volatile matter are achieved within a narrow range of O:C ratios between 1.05-1.13, whilst the least reactive coal with high fixed carbon achieves its maximum CGE value at a higher O:C ratios of 1.36. Importantly, the model is able to reflect the significant differences in gasification behaviour of the four coals, which is consistent with lab-scale and larger-scale investigations. This work demonstrates the relevance of bench-scale gasification data in the assessment and interpretation of coal gasification behaviour under complex high pressure and high temperature conditions using appropriate mechanisms and sub-models. (C) 2015 Elsevier B.V. All rights reserved.