Combustion behavior of xylite/lignite mixtures D. Vamvuka * , E. Kastanaki, M. Lasithiotakis, C. Papanicolaou Department of Mineral Resources Engineering, Technical University of Crete, Kounoupidiana, 73100 Chania, Greece Received 23 May 2003; accepted 4 November 2003 Abstract The behavior and the kinetics in nitrogen and air of two low-rank coals (lignite and xylite) and their blends, as well as the compatibility of the component coals in the blends were evaluated, in an effort for the rational use of poor coals. The experiments were conducted in a thermobalance system, at non-isothermal heating conditions, with heating rates of 20 and 100 °C/min, in the temperature range of 25–850 °C. Material particle size was )100 lm. A first-order parallel independent reactions model and a power law model fitted successfully the rate data of pyrolysis and combustion, respectively. Activation energy values and reaction orders ranged from 23 to 182 kJ/mol and 0.8 to 2 respectively. The heating rate did not affect the kinetic parameters considerably, however when this was increased the reactions were shifted to higher temperatures and the rates were greater. The pyrolysis kinetics of lignite/xylite blends could be sufficiently predicted, based on the data of the individual fuels. However, this was not true in the case of char combustion. Blending of lignite with xylite, in any proportion, seemed to cause some interactions between the component coals in air. Ó 2003 Elsevier Ltd. All rights reserved. Keywords: A. Coal; B. Pyrolysis, combustion; C. Thermal analysis; D. Reaction kinetics 1. Introduction The problem of the rational use of low-rank coals has not yet been solved. In Greece, where lignites constitute the major energy resource, covering 70% of the demand for electricity production, this problem is caused by the significant share of non-combustibles and their un- favorable chemical properties, which results in a lower availability and a higher running cost of the plants. An efficient way of solving the above problem is probably blending. Coals can be blended to produce a usable fuel from the coals available and increase the number of supply options. Also, apart from reducing plant costs, they can be blended to improve the com- bustion performance and meet emission limits [1]. However, as coal quality affects virtually every compo- nent in the power plant, the impact of blending different coals is not known and blending can result in serious problems. Since reaction rates are influenced by the composition and the particle size of the coals, as well as the operating conditions and plant design, the knowledge of the re- activities and the kinetics of the blends will provide valuable information for the proper operation of the combustion systems. The reactivities of carbonaceous materials are commonly investigated by using iso- thermal and dynamic thermogravimetric experiments. There is an extensive literature on coal pyrolysis and combustion kinetics [2–7], however this is limited as concerns the kinetics of the blends [6,8]. Over- simplifications, such as the description of the whole process by a pseudo-first-order reaction, have been widely used [3,4,7,9], however more sophisticated mod- els, describing the pyrolysis/combustion process as a series, consecutive or parallel first/nth-order processes, are being developed recently, as being more accurate [5,10–12]. The overall aim of this work was to evaluate the compatibility of the component coals in the blends, with respect to pyrolysis or combustion and investigate pos- sible interactions, which may or may not be beneficial, in an effort for the rational use of poor coals. Thus, the devolatilization and combustion reactivities of a lignite and a xylite and their blends were studied under non- isothermal thermogravimetric experiments and kinetic models, which could approximate reactivity changes * Corresponding author. Tel.: +30-821-37603; fax: +30-821-69554. E-mail address: vamvuka@mred.tuc.gr (D. Vamvuka). 0008-6223/$ - see front matter Ó 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.carbon.2003.11.002 Carbon 42 (2004) 351–359 www.elsevier.com/locate/carbon