FINAL TECHNICAL REPORT

September 1, 1997, through August 31, 1998

Project Title: A Novel High Efficiency Enhanced Gravity Separation Using Dense Medium

ICCI Project Number: 97-1/2.1A-1

Principal Investigator: R. Q. Honaker, Southern Illinois University

Other Investigators: B. G. Rajan, and N. Singh

Southern Illinois University

Project Manager: K. Ho, ICCI

ABSTRACT

The main objective of the project was to develop a highly efficient enhanced gravity separation (EGS) process for cleaning fine coal (16 x 325 mesh) using dense medium. Extensive test work was conducted on both dense medium (D.M.) and water-only EGS processes using the Falcon Concentrator as the EGS unit, two easy-to-clean coal samples and two difficult-to-clean middling samples as feed. Statistically designed experiments were conducted to optimize the performance of the dense medium and water-only Falcon processes. Under optimum conditions, the D.M. Falcon process provided a clean coal yield of 78.8% with 5.7% ash and 2.9% total sulfur for an Illinois No. 6 seam coal containing 15.5% ash and 3.8% total sulfur. The pyritic sulfur content was reduced from 1.2% to 0.6%. For another easy-to-clean Illinois No 6 seam coal with 17.5% ash and 1.5% total sulfur, the D.M. Falcon process achieved a clean coal yield of 78.9% with product ash and total sulfur contents of 5.7% and 1.0%, respectively. According to feed washability analysis data for both coals, these separation performances equate to an organic efficiency of nearly 99% and a 7% weight unit improvement in clean coal yield compared to the optimized water-only Falcon process results. Treatment of an Illinois No. 5 middling sample by the D.M. Falcon process reduced the feed ash and total sulfur contents from 22.3% and 2.5% to 7.5% and 1.6%, respectively, while recovering 62.2% of the feed mass to the product. Another Illinois No. 5 middling sample having 29% feed ash was cleaned to 6.9% ash with 51.6% yield in the D.M. Falcon. The feed total and pyritic sulfur contents were reduced to 1.5% and 0.8% from a feed of 2.8% and 1.7%, respectively. For the difficult-to-wash middling samples, the D.M. Falcon provided a 16% weight unit improvement in clean coal yield over the water-only process and achieved an organic efficiency greater than 90%. The water-only and D.M Falcon processes were also able to produce Phase I compliance coal from the middlings samples, which had an estimated SO₂ emission greater than 4.5 lbs/MBTU in the feed. Performance simulations conducted for the Illinois No. 5 plant indicate a 4.1% increase in clean coal yield at current target product quality levels by replacing the existing spiral circuit with D.M. Falcon. Preliminary economic analysis revealed that the clean coal yield improvement translates to a $4.2 million increase in annual income after discounting the capital cost invested at a rate of 12% per annum. The combination of a high "g" force and dense medium seems to provide excellent separation efficiency for cleaning the 16 x 325 mesh particle size coal.

EXECUTIVE SUMMARY

The major goal of this project was to develop a highly efficient, high capacity enhanced gravity concentration process using dense medium for treating the -28 mesh particle size fraction in coal preparation plants. The test program evaluated a novel application of magnetite-based heavy media in a continuous enhanced gravity separator (EGS) commercially known as the Falcon Concentrator.

Funding provided by the ICCI and Illinois coal companies has facilitated test programs to be conducted by the principal investigator which have revealed that relatively low D₅₀ values of about 1.6 can be achieved on the 28 x 325 mesh particle size fraction using water-only enhanced gravity separation technologies. This separation was achieved in a full-scale Falcon Concentrator treating 75 tons/hr of feed at a volumetric feed rate of 1800 gallons/minute. However, the corresponding E_p values are between 0.12 and 0.18, which are comparable to those associated with spiral concentrators. Though these E_p values may represent an acceptable efficiency at the high gravity cut-points, gravity cut-points lower than 1.6 typically results in significant loss in mass yield due to the increase in the amount of near-gravity material and the relative process inefficiencies. Thus, a unit operation providing an even lower D₅₀ value than the 1.60 value and a higher separation efficiency is needed to realize the yield optimization goal of the proposed project.

Previous studies conducted on the Falcon Concentrator have found that a significant amount of coal in the 28 x 100 mesh particle size fraction is lost to the tailings. Though the high g-force provided by the Falcon concentrator enhances the separation efficient of the finer particle size fractions (100 x 325 mesh), it also leads to faster settling of the coarse coal into the particle bed, which reports to the tailings process stream. It is believed that, by using a dense medium in the Falcon Concentrator, the stratification of the particle bed is more controlled and the effect of particle size during the separation process is greatly reduced. Preliminary feasibility test results validated this hypothesis.

The proposed project objective was only to test 28 x 325 mesh particle size coal. However, installation of new underflow nozzles in the C10 Falcon Concentrator, which eliminated the plugging problem realized during a previous investigation, allowed successful testing of a wider particle size range, i.e., 16 x 325 mesh. The new nozzles provide a variable diameter, for which, the size is controlled by compressed air.

Four different coal samples were treated in both dense medium and water-only Falcon processes. Coals No. 1 and 2 were collected from operating coal preparation plants treating coal from the Illinois No. 6 seam. Based on washability analysis data, both coals were characterized as easy-to-clean with Coal No. 1 being a high sulfur coal. Coals No. 3 and 4 were collected from the middling stream of spiral concentrators installed at a preparation plant treating Illinois No. 5 seam coal. Both middling sample sources are currently waste streams. Coals No. 3 and 4 contained a significant amount of near gravity material and, thus, were characterized as difficult-to-clean. Therefore, effective treatment of the two middling coals would indicate a high level of efficiency for any gravity-based process.

Experiments based on a Plackett-Burman design were conducted on the dense medium Falcon operation to identify the most significant process variables. Based on these results, four important process variables were identified for a more detailed test program conducted according to a Box-Behnken design. Separation performance results were used to evaluate the effect of the process variables and optimize the performance of the dense medium enhanced gravity separation. Statistically designed experiments were also conducted to optimize the performance water-only Falcon process.

Table 1 shows the optimum performances achieved for the two easy-to-clean coals for the dense medium (D.M.) and water-only (W.O.) Falcon processes. The results indicate that the dense medium process provides a superior separation performance resulting in a 7.8% weight unit increase in clean coal yield over the water-only results at the given product quality level. Optimization validation experiments confirmed the reproducibility and the performance of the dense medium operation. A comparison with feed washability data indicated that the dense medium process provided an organic efficiency nearly 99%. The partition curves obtained for the dense medium Falcon process indicated that a density cut point of 1.41 was achieved with a low probable error (E_p) value of 0.038 compared to that of 0.075 normally achieved in conventional fine coal cleaning using dense medium cyclones.

Table 1. Optimum separation performance achieved in dense medium and water-only Falcon processes for treating Coals No. 1 and 2.

Coal No. 1

Coal No. 2

	D.M.	W.O.	D.M.	W.O.
Feed
Ash (%)	15.5	14.8	17.5	17.4
Total sulfur (%)	3.8	3.8	1.5	1.5
Pyritic sulfur (%)	1.2	1.2	0.6	0.6
BTU/lb lbs SO2/ MBTU	11386 6.67	11425 6.65	11104 2.70	11198 2.68
Product
Ash (%)	6.2	6.3	5.7	5.7
Total sulfur (%)	2.9	3.1	1.0	1.0
Pyritic sulfur (%)	0.6	0.8	0.2	0.2
BTU/lb	13586	13312	13989	13958
lbs SO2/ MBTU Clean coal yield (%)	4.22 78.8	4.67 71.0	1.43 78.9	1.43 71.2
Organic efficiency (%)	99.2	88.2	98.9	89.2
Density cut-point, d50	1.41	1.48	1.42	1.53
Probable error, Ep	0.038	0.105	0.040	0.110

The dense medium process was also able to achieve high quality products, i.e., low ash content, with reasonable clean coal yield values. For example, a product containing 3.9% ash was achieved while recovering 62.7% of the feed material to the product. In comparison, the ultimate froth flotation performance as predicted by the Advanced Flotation Washability (A.F.W.) indicates only a 36.7% yield to the product at the same product grade.

The optimum performances of the dense medium and water-only Falcon processes treating the more difficult-to-clean Coals No. 3 and 4 are summarized in Table 2. The dense medium process performed well with an organic efficiency above 90%. Reflective of the high efficiency of the D.M. process and the difficult-to-clean characteristics, the improvement in clean coal yield of 17% weight units was significantly higher than the differential obtained for Coals No. 1 and 2. The dense medium process provides a more effective stratification of the near gravity material and reduces misplacement, which results in higher clean coal yield values at the same product quality levels.

Table 2. Optimum separation performance achieved in dense medium and water-only Falcon processes for treating Coals No. 3 and 4.

Coal No. 3

Coal No. 4

	D.M.	W.O.	D.M.	W.O.
Feed
Ash (%)	22.3	22.3	29.0	29.0
Total sulfur (%)	2.5	2.5	2.8	2.8
Pyritic sulfur (%)	1.7	1.7	1.7	1.7
BTU/lb lbs SO2/ MBTU	10691 4.68	10691 4.68	9593 5.38	9593 5.38
Product
Ash (%)	7.5	7.8	6.9	8.1
Total sulfur (%)	1.6	1.9	1.5	1.5
Pyritic sulfur (%)	0.4	0.6	0.8	0.8
BTU/lb	13190	12696	13394	13033
lbs SO2/ MBTU Clean coal yield (%)	2.18 62.2	2.5 45.3	2.32 51.6	2.22 32.8
Organic efficiency (%)	91.3	66.6	93.8	60.6
Density cut-point, d50	1.43	1.55	1.49	1.55
Probable error, Ep	0.048	0.128	0.055	0.178

Additional experiments were conducted on a Humphrey spiral concentrator using Coal No. 4 as feed to compare the performances of dense medium and water-only enhanced gravity separation processes with the conventional fine coal cleaning operation. Due to inherent inefficiencies involved in the spiral process, the product ash content could not be reduced below 18.9%, which was achieved while yielding 61.9% of the feed mass to the product. The total and pyritic sulfur contents were reduced only to 2.3% and 1.3%, respectively. The inability of the spiral process to achieve density cut points below 1.8 is the main reason for the poor performance. At a comparable product ash level (18.9%), the clean coal yields obtained from the dense medium and water-only Falcon processes were 74% and 65%, respectively. It may also be noted from Table 2 that both dense medium and water-only processes were able to produce Phase I compliance coal from the middling samples.

A preliminary study evaluating the effect of magnetite particle size distribution on the performance and expected loss of medium from the dense medium Falcon process was conducted. Experiments conducted using fine (d₈₀ = 17 microns ) and coarse ( d₈₀= 31 microns) magnetite for treating Coal 4 showed that process efficiency is not affected by the use of coarse magnetite. The dense medium recovery studies were conducted using a wet drum magnetic separator arranged in rougher-cleaner-scavenger circuit. The coal slurry with the fine magnetite as dense medium was used in the media recovery studies to provide the worst case scenerio. The media loss was found to be about 3 kgs/ton of raw coal treated. The use of the coarse magnetite will reduce the loss without significantly affecting the dense medium Falcon process efficiency.

Using the performance data obtained from the dense medium Falcon process, computer simulations were conducted to evaluate the benefits expected for the overall coal preparation plant operation. Computer programs and data generated during a previous ICCI project investigation were used to optimize the existing plant operation and evaluate the improvement achieved by replacing the conventional fine coal cleaning devices with the dense medium Falcon process. The processing plant evaluated in this study treats Illinois No. 5 seam coal in four circuits consisting of a dense medium bath, cyclones, spirals and conventional froth flotation. At the current product quality levels of 8% ash and 1.1% total sulfur, replacing spirals by the dense medium Falcon provided a 4.1% weight unit improvement in overall plant clean coal yield. A preliminary economic analysis was also conducted to evaluate the monetary benefits of using the dense medium Falcon operation. The requirements for additional capital cost and recurring costs such as dense medium loss were taken into consideration. The economic analysis predicted a $4.2 million increase in annual income due to the dense medium enhanced gravity separation after discounting the capital cost invested at a rate 12%.

The extensive evaluation conducted in this investigation indicates that the dense medium Falcon operation is a unique high efficiency and high capacity process for treating coal over a wide particle size range, i.e., 16 x 325 mesh. Apart from the techno-economic benefits, the dense medium Falcon may also provide operational flexibility such as the ability to achieve a wide variation in separation density cut points (d₅₀) without changing the feed medium density. For example, a range of density cut points from 1.4 to 1.6 was achieved by manipulating bowl speed and underflow valve air pressure while maintaining the feed medium density at 1.5. The dense medium Falcon process is also amenable to on-line process control to achieve optimum process and plant efficiency while producing target product qualities.