The transient swelling behaviour of large (-20+16 mm) South African coal particles during low-temperature devolatilisation

In order to predict coal behaviour during certain utilisation processes, and to improve and develop coal conversion technologies, it is important to understand the swelling behaviour and plastic properties of coal. X-ray CT (computed tomography) and mercury submersion were used to quantitatively and qualitatively describe the swelling behaviour (N-2 atmosphere, 7 K/min) of large particles (-20 + 16 mm) of South African coals, and were compared to conventional swelling related characteristics for powdered coals. Three South African coals, with low, intermediate and high Free Swelling Indices (FSI) were selected. Single coal particles were devolatilised (from ambient to 700 degrees C at 7 K/min) in a furnace equipped with a video camera. Image analysis was conducted on the images obtained from the furnace, to describe the swelling behaviour of single coal particles as a function of devolatilisation temperature. Xray computed tomography (X-ray CT) and mercury submersion measurements were used to analyse particles before and after devolatilisation, to quantify the degree of swelling. While the FSI, dilatometry and Gieseler results did not indicate any plastic behaviour for the TWD sample (-212 mu m particles), the large particles did exhibit a significant degree of swelling during devolatilisation. The GG and TSH large coal particles started swelling at a slightly lower temperature than the initial softening temperature obtained by dilatometry for the -212 mu m coal particles. Image analysis qualitatively described the swelling behaviour of large coal particles during devolatilisation, while both the X-ray CT and mercury submersion methods quantified the volumetric swelling ratio of large coal particles. The average swelling ratios obtained for TWD, GG, and TSH respectively were 1.9 +/- 0.5, 2.1 +/- 0.6 and 2.5 +/- 0.9 from image analysis and 1.8 +/- 0.4, 2.2 +/- 1.0 and 2.5 +/- 1.0 from mercury submersion. The anomaly found for the TWD coal is expected, since mass transfer of tar and volatiles out of the particles will influence the thermoplastic behaviour, and consequently the difference in swelling behaviour due to particle size variation. Therefore, it is hypothesised that the FSI, and other conventional techniques used to describe the plastic behaviour of small particles of coal, can in general not be used for the prediction of large coal particle swelling. (C) 2014 Elsevier Ltd. All rights reserved.