The dry beneficiation of South African small coal in a dense medium fluidized bed
Conventionally, wet processing was preferred to beneficiate coal based on its sharp separation efficiency; however, the excessive process water constraint is an extreme drawback and proves that wet beneficiation is currently an unsustainable coal washing technique in some parts of the world. Therefore, research into the development of dry beneficiation technologies, especially in coal-rich, semi-arid countries like South Africa is strongly motivated. In principle, dry technologies are similar to most wet beneficiation techniques that are currently industrialized. These operate based on the relative movement of a particle within a medium in accordance to the difference in said particle and medium density. In the simplest terms, particles denser than the medium sink and those less dense, float. The preferred method of dry dense medium separation is proving to be the air dense medium fluidized bed (ADMFB) technology. In the ADMFB, air at specific velocity is blown through a particulate bed of suitable dense medium. This forms a fluidized bed of particles with pseudo-fluid properties such as, and most applicable to this study, density. Upon the addition of coal to the suspended medium, stratification occurs and the heavier coal particles (associated with high mineral content and low calorific value) sink to the bottom of the bed, and in turn the lighter, better quality particles float. Experimentally, a mixture of coal and medium is loaded into a column and fluidized with air for a period of time. After completion the particulate suspension is allowed to settle into to a packed bed state which is cut in layers of specific height and sampled for analyses. The extent of separation is observed from the quality of the coal in terms of the ash yield, calorific value and density, reporting to each layer. The purpose of this study was to determine whether ADMFB is suited for the effective beneficiation of South African small (+5.6-13.2mm) coal particles. The separation efficiency of the fluidized bed, (how effectively high-density gauge and low-density coal are separated in the ADMFB), was evaluated by considering the following objectified variables: the effects of differing particle size distributions, variance in medium-to-coal ratios and the influence of activated vibration. Auxiliary investigations toward the quality of the feed coal were also conducted to further comprehend the significance of this variable on the separation efficiency of the fluidized bed. Another major outcome of this study was to design an enlarged ADMFB for processing of large particle sizes and increased bed loads. This was approached by an initial fundamental process design in which the modified Ergun and associated fluidization equations were used to obtain the minimum fluidization velocities and corresponding bed pressure drop and height for each variable considered. From these results, the bed geometry and dimensions of the new assembly were decided upon. A 0.3m by 0.3m square bed structure was selected, made up of eight 0.05m transparent PVC layers that fit perfectly into another to prevent air and material leakages. This makes up the fluidizing layer of the bed, which is mounted on an air distributer mechanism that is affixed to a centrifugal fan delivering an even distribution of air at the required velocity. The effect of vibration is of interest to this study and bed vibration, when required, is induced by means of an oscillating vibratory motor with adjustable frequency and amplitude. Accurate flow and pressure readings, required for establishing a stable fluidized bed, were attained from an air velocity sensor and water based manometer, respectively. Dry beneficiation of the coal was attained within the ADMFB, in varying degree, for each of the variables listed above. For a theoretical product of 75% bed volume (or the top 3 bed layers), ash yields obtainable are recorded to vary between 13 and 17.6% with a corresponding mass yield percentage range of 27.6-34.2. For the experiments conducted in this study, these values worsen slightly when considering the addition of dense medium and vibration. When focussing on the ash values in the feed, top and bottom layers of the bed, a decrease in coal particle size distribution (PSD) proves to have an effect on the extent of separation in the bed. All considered PSD ranges yielded comparable ash values in the top, intermediate and bottom bed layers but the larger PSD (+11.2-13.2mm) yielded the best performance curves when considering ash value and mass yield. Ash content values ranging from 18-30%wt were obtained in the top bed layers over all variables considered, depending on the ash content of the feed and whether dense medium or vibration are present during operation. It was further established that the addition of dense medium (magnetite) did not have a remarkable effect on the separation efficiency of the ADMFB. The experimental runs conducted with dense medium yielded less desired results in the top, intermediate and bottom layers of the bed when compared to those of only coal. The activation of vibration too, showed no improvement to the extent to which the ADMFB separates the coal, middling and gangue products. However, noticeable was that a decrease in PSD, addition of dense medium and activation of vibration had a significant impact on the operability of the ADMFB in terms of minimum fluidization requirements and bed stability. Coal with considerable amounts of intrinsic ash forming minerals, which cannot be liberated by means of crushing, prove extremely difficult to beneficiate. Conclusively, it was established that the quality of the coal in terms of amount of inorganic matter (minerals) present and the measure to which these are intimately mixed with the organic matter (microlythotypes), significantly influenced bed segregation. From this it can be noted that a blended feed coal, containing a variety of qualities, is best suited for the dry beneficiation of coal in an ADMFB and that extensive optimization of the operation in all aspects is crucial. Furthermore continuous operation of the bed is required to ensure that the drawbacks of a batch-type experimental run, most especially with regard to vibrated dense medium beds, are negated.
- Engineering