Coal Preparation » Dewatering
High moisture content in export coals increases transport and handling costs and invokes contract penalties. Considerable effort has been focussed on fine coal dewatering, but significant variation in the moisture content of the coarse product is also important. Very little work has been carried out to determine the magnitude and causes of these variations. ACARP supported this AMIRA project which was established to investigate the problem.
The moisture content of coarse coal after centrifugation has been found to vary significantly:
The Grubbs estimation technique has shown that the MCI Model TBM101 on-line moisture meter is capable of providing accurate and precise (plus/minus 0.3%) information in real plant environments. It is not significantly affected by the presence of magnetite or thin (about 65mm) beds of coal. The meter is capable of maintaining its calibration over significant time periods (a number of months).
Normal distributions or multiple normal distributions for multiple seam mines, appear to provide a convenient method of describing the plant, fine coal and coarse coal moistures. This could be of use in determining real changes to plant operations and the application of statistical process control procedures.
Little dependence of coarse coal moisture was found with respect to time of plant operation. There was a small positive correlation with the feed rate to the plant. The product ash value for the two coals examined also had little impact.
Laboratory centrifuge technique has been developed, which for relatively coarse coal feeds, has identified that there are three types of water associated with a coal with respect to its potential for removal by centrifugation:
There is an amount of water removed easily with centrifugation. This excess water appears to have little relationship with the coal.
There is an amount of water which is closely related to the coal surface, causing a slower rate of removal.
There is a Non-Centrifugable Moisture (NCM) which cannot be removed by centrifugation even using much longer times. It comprises a small amount of surface water, as well as an amount of internal water within the coal structure. Particular coals have different capacities for holding internal water.
It has been found that the results from the laboratory centrifuge are erroneous when the samples have been allowed to begin drying. This is considered to be due to problems of rewetting the coal, particularly with respect to water being unable to penetrate the dried internal pore structure of the coal.
The Non-Centrifugable Moisture results from the laboratory centrifuge work can be interpreted as the coal having an internal moisture, NCM i , (which is unaffected by particle size) and an external moisture, NCM s , which varies with particle size, and is proportional to the surface area. Values of NCM i and NCM s are given for the coals tested.
Four boreholes were obtained form the coal seams involved in the plant investigations. There are significant differences between the NCM i values, with the Goonyella Middle seam having a substantially higher value (4.2%) than the Goonyella Lower (3.1%) and the Middle seam from the Isaacs Pit (3.4%). All the Goonyella cores are much higher than that for the Dysart seam at Norwich Park (2.3%). The surface moistures were similar for all the material from the boreholes.
The mercury porosimetry tests provide relative internal pore volumes for the different coals tested. All the Norwich Park samples have lower internal porosities than those from Goonyella. For both coals, the dull material has a higher pore volume than the bright, and also retains more mercury during extrusion. The dull coal from the Goonyella Middle seam contains significantly higher porosity than the Lower dull and all the bright samples. The dull material from the Isaacs pit had a porosity between the Middle and Lower seams.
Analysis of the results from the major samples obtained from the two plants, as well as samples obtained at five other mine sites has indicated that the moisture level of the coal after centrifugation and its propensity to vary, is controlled by:
- the amount of moisture held internally within the coal;
- the specific surface area of the coal being presented to the centrifuge;
- the hydrophobicity of the coal as indicated by the rank (reflectance);
- the amount of ultra-fine (slimes) material associated with the coarse coal, which appears to affect the drainage of water from the coal particle surface.
The overall range of the rank of the coals involved in this investigation, as measured by reflectance, varied from 0.7 to 1.65. The model was able to explain 80% of the variation in the 25 samples investigated.
At all sites investigated, it was found that there was a significant amount of coal breakage in the centrifuge. Evidence suggests that the majority of the breakage occurred as the coal leaves the centrifuge, and any analysis based on the final product size distribution could be erroneous.
At Goonyella, washing the coal (ie the removal of the ultra-fine particles led to decreases in moisture level.
Scanning electron microscopy has identified that the surface of the coarse particles is covered by a layer of ultra-fine particles, with the majority of these being carbonaceous in nature.
A model incorporating the above factors has been developed. It provides an approach for estimating the moisture level which can be achieved by centrifugation of coarse coal.