Coal Preparation » Dewatering
The objective of this project was to determine the potential to effect a reduction in fine coal product moisture using a novel agglomeration technology. Following agglomeration, the final product was subjected to pressure filtration tests at both 1 and 7 bar, and the final moistures determined. Initial experiments in the present study failed to show the expected level of benefit from the agglomeration, with the higher than expected moistures believed to be a result of the relatively fine size distributions, defined by the Sauter mean. Previous work had involved conventional flotation product consisting typically of coarser coal (predominantly greater than 20 microns) and entrained ultrafine mineral matter (less than 20 microns), hence the agglomeration recovered the coal only, rejecting the ultrafine mineral matter, resulting in a relatively coarse product, and hence delivering remarkably low moisture following the filtration of the agglomerated product.
The feed sourced for this project was typically derived from the thickener feed of a given coal preparation plant, hence was relatively fine compared to typical flotation product. The Reflux Flotation Cell (RFC) was in turn applied to this thickener feed, ahead of the novel agglomeration, the purpose of the agglomeration being to lower the final product moisture. The RFC delivered a product dominated by ultrafine coal, reflecting the size distribution of the feed, and due to its extreme cleaning, a product free from ultrafine mineral matter. Thus, when the agglomeration was applied to the RFC product, there was effectively no change in the size distribution, because there was effectively no mineral matter to reject, hence the reduction in moisture, while significant, was not as great as expected.
A systematic approach was taken to evaluate the effect of the agglomeration on the fine coal moisture. Model feed materials were generated using a binary approach based on combinations of the clean RFC product and the ‐0.038 mm RFC tailings. Other model feeds were formed by classifying the RFC product to remove the ultrafine portions of the clean coal, and then again introducing portions of the ‐0.038 mm RFC tailings.
This revised approach provided a benchmark for assessing the factors governing the final product moisture. The first series of filtration experiments were conducted on these model feeds with no agglomeration or oil addition. In general, the results of these pressure filtration experiments showed a strong correlation between the particle size and both the final product moisture and breakthrough times. The base case RFC product (no ‐0.038 mm tailings) had a moisture of 34 wt%. Interestingly, the introduction of ultrafine ‐0.038 mm mineral matter resulted in similar moistures, however the breakthrough times increased exponentially to unacceptable levels meaning satisfactory moisture could not be achieved using equipment of finite capacity. This work was extended to the removal of ‐0.020mm, ‐0.038mm, and ‐0.045mm RFC product, leading to a minimum moisture of 5.5 wt% on the +0.045 mm portion at a filtration pressure of 1 bar. Varying levels of ‐0.038 mm tailings were also introduced, leading to higher moistures and breakthrough times. Overall, the breakthrough times were found to decrease from 10 minutes to a few seconds through the removal of the ultrafine particles. Further improvement was achieved in all cases when the filtration pressure was increased to 7 bar.
Agglomeration of the RFC product resulted in measurable reduction in the product moisture and breakthrough time. Moistures were reduced from 34 wt% (no binder) to 24 wt% (with binder), while breakthrough times were reduced from 10 minutes to 10 seconds by using the agglomeration, and a filtration pressure of 1 bar. Moistures as low as 18 wt% were achieved using the agglomeration and a filtration pressure of 7 bar.
Conventional flotation product was simulated by combining the RFC product screened at 0.020 mm with portions of the ‐0.038 mm mineral matter. Following agglomeration, and then filtration, a moisture of 19 wt% was obtained, largely due to the removal of the ‐0.038 mm mineral matter by the agglomeration. It is noted that when the agglomeration was applied to the +0.020 mm RFC product, with no ‐0.038 mm mineral matter added, a lower moisture of 14.6 wt% was obtained. These results show major improvements in product moistures and breakthrough times following the agglomeration, especially in the absence of the ultrafine coal and mineral matter below 0.020 mm.
The method of organic liquid addition was found to impact the final product moistures and breakthrough times, with the SMO (emulsifier) and kerosene oil mixture performing equally to the novel agglomeration binder. A clear trend of decreasing product moisture and breakthrough times was also found as the organic liquid dosage increased, covering doses of up to 10 wt%. It is noted that there was no beneficiation involved here, as all the material was sent to the filtration system. It is concluded that a dose of about 1‐2 wt% organic liquid delivered the greatest moisture reduction per unit of oil addition. The oil needs to include an emulsifier to help stabilise the oil drops, and in turn assist in the delivery of the oil to the particles.
Overall, the lowest product moisture and breakthrough time is achieved through the formation of clean coal product containing (i) negligible slimes by using a strong positive bias flux in the flotation (via the RFC) (ii) limiting the presence of ultrafine coal through fast processing (exploiting the poorer flotation kinetics of the ultrafines) (iii) adding about 1‐2 wt% organic liquid (oil and emulsifier) and (iv) applying an elevated pressure of about 7 bar to the filtration. It is important to recognise the contributions of all of these factors in order to obtain the best possible result. An economic analysis based on the interplay between moisture reduction, lower yield, reagent costs, and higher pressures would be needed on a case‐by-case basis.