United States Environmental Protection Agency Air and Energy Engineering Research Laboratory Research Triangle Park NC 27711 Research and Development EPA/600/S7-85/032 Apr. 1986 &EPA Project Summary Economic Evaluation of Oil Agglomeration for Recovery of Fine Coal Refuse L Larkin and J. D. Maxwell In this project economics of an oil- agglomeration process with and with- out an oil recovery system were evalu- ated for recovering coal fines from a fine refuse stream of 105 ton/hr* from a coal preparation plant. The two base case processes studied are an oil- agglomeration process in which heptane is used and recovered and an oil- agglomeration process in which fuel oil is used and blended with the product. The economics for both processes were estimated with and without a pond credit (savings in coal preparation plant investment resulting from the smaller waste disposal pond needed for the oil- agglomeration process). The total capi- tal investments for the recovery and nonrecovery processes without a pond credit are $21 million and $13 million, respectively. With the use of the pond credit, the total capital investment for the recovery process is $9 million, and a capital investment credit of $0.2 million is received for the nonrecovery process. The first-year annual revenue require- ments for the recovery and nonrecovery processes are $6.4 million (0.86 $/106 Btu) and $8.0 million (1.10 $/10s Btu) with a pond credit and $8.5 million (1.15 $/10e Btu) and $10.3 million (1.42 $/108 Btu) without a pond credit, respectively. These costs compare quite favorably with an eastern bituminous coal which has a heating value of 11.000 Btu/lb and cost of 1.85 $/10B Btu (40.70 $/ton). Both the recovery and the nonrecovery processes appear to be economically feasible, but the •Readers more familiar with metric units may use the conversion factors at the back of this Summary. recovery process is more cost-effective for recovering fine coal from refuse streams. This Project Summary was developed by EPA's Air and Energy Engineering Research Laboratory. Research Triangle Park, NC. to announce key findings of the research project that is fully docu- mented in a separate report of the same title (see Project Report ordering in- formation at back). Introduction In this report, the economics of an oil- agglomeration process with and without oil recovery system are evaluated. The two base case processes studied are an oil-agglomeration process using heptane with a heptane recovery system and an oil-agglomeration process using fuel oil that does not have an oil recovery system. The design data for the processes are based on information from vendors and researchers of oil-agglomeration and oil recovery systems. The econmics of both processes are presented and compared, and several case variations are also examined. Background Much of the coal lost in the waste from coal-cleaning plants consists of fine coal that is difficult to separate from noncoal minerals with conventional coal-cleaning techniques. The use of highly efficient coal-cleaning processes and cleaning of coal with finely dispersed pyrite, both of which necessitate more extensive use of fine coal-cleaning techniques, can dra- matically increase the quantity of coal lost in the fine waste stream and the ------- volume of fine waste produced. The coal loss represents a substantial potential resource, and the waste itself, which consists of a slurry that is difficult to dewater and must be confined in a pond, poses increasingly serious economic and environmental problems. Consequently, interest in methods of cleaning and recovering fine coal has increased. One of the most effective methods is the oil- agglomeration process developed by the National Research Council of Canada. The oil-agglomeration process is a means both of cleaning fine coal and of recovering the coal in a more useful form. Fine coal (in the minus 28-mesh range) is dispersed in an agitated vessel with a light oil such as heptane, fuel oil, or kerosene, and in some cases a small quantity of a binder such as asphalt is used. By carefully controlling the degree of agitation, the oil is dispersed in small droplets in which the coal and other oleophilic (oil attracted) minerals collect, forming small spherical particles. Over 90% of the coal in a fine coal slurry can be recovered in this manner. In spite of its effectiveness, the process has not gained commercial acceptance because of the cost of the oil used. The agglomeration process becomes more economically at- tractive when designed for low oil-to-coal ratios; however, there is also a corre- sponding decrease in coal recovery efficiency and product quality. Recovery and reuse of the oil would allow the agglomeration process to operate at higher oil consumption levels with greater coal recovery and improved product quality. The recovery process is ostensibly simple: the mechanically dewatered ag- glomerate is heated to vaporize the oil and the vapor is condensed to recover the oil. Several companies have investigated aspects of an oil-agglomeration process with an oil recovery system but many technical and practical details remain to be defined. The primary technical chal- lenge is the development of a heating system that provides efficient and con- trollable heat transfer without thermal and mechanical damage to the particles. Design and Economic Premises The design and economic premises used in this study were developed by the Tennessee Valley Authority (TVA) for economic comparisons of processes re- lated to coal-cleaning and emission con- trol in electric utility applications. The conceptual process designs are based on information provided by vendors of oil- agglomeration and oil recovery equipment and systems. The plant is assumed to operate at 5,500 hr/yr for 30 years, with a total operating life of 165,000 hours. The quantity and composition of the feed to the oil-agglomeration plant for this study are assumed to be similar to the fine coal refuse produced by the Brecken- ridge Camp No. 11 coal-cleaning plant near Breckenridge, Kentucky. The refuse consists of a 10% solids slurry produced at a rate of 5.8 million ton/yr. The solids have a maximum size of 28 mesh and consist of 47.5% coal, 2.5% pyrite, and 50% other noncoal minerals. The base case design conditions for the oil- agglomeration process (using heptane) with a heptane recovery system (recovery process) and the agglomeration process (using No. 2 fuel oil) without a recovery system (nonrecovery process) are shown Table 1. Process Design Conditions in Table 1. Heptane was selected as the oil in the recovery process because of its distinct properties which make it easier to recover than other oils, and No. 2 fuel oil was used in the nonrecovery process mainly because of its lower cost. A 30-day-capacity holding pond and a 30-year-capacity waste disposal pond are provided for the oil-agglomeration plant as a replacement for the large volume waste disposal pond that would have been required for the coal-cleaning plant if the waste had not been processed. The waste disposal ponds are square earthen- diked impoundments with a median di- verter dike and a 12-in. clay lining. Raw materials consist of a propane precipitated asphalt at 189.2 $/ton and commercial grades of heptane. No. 2 fuel oil, and kerosene at 1.60, 1.09, and 1.32 $/gal., respectively. All raw materials are Process Heptane With Recovery Fuel OH Without Recovery Feedstock Rate. 10*lb/hr Slurry concentration. % solids Solids composition. %by wt(dry) Coal Noncoal minerals Pyrite Operating Conditions Time, hr/yr Coal recovery. % of coal feed Oil. % of undried product Asphalt, % of undried product Oil recovery. % of oil feed Product" Production, ton/yr Solids composition. % by wt (dry) Coal Noncoal minerals Sulfur Water Heating value. Btu/lb (dry) Waste Rate. W'lb/hr Solids, % Coal. % of waste 2.1 10 47.5 50 2.5 5.500 92 18' 2 97.7 289.000 89 8 3 <5 12,900 1.9 5 0.3 2.1 10 47.5 50 2.5 5,500 90 6.1" 0 0 302.000 83 14 3 30 12.000 1.9 5 0.4 * This number corresponds to 14.1% of oil based on the weight of the feed solids and 21.4% oil based on the dry weight of the agglomerated product (water-free basis). "This number corresponds to 5.0% oil based on the weight of the feed solids and 8.7% oil based on the dry weight of the agglomerated product (water-free basis/. CAII product percentages and the heating values are based on the dry weight of the product (coat and all noncoal minerals, including pyrite) and do not include the weights or effects on heating value ofresidualoH. asphalt, or water (the heating values of the residual oil and asphalt ate taken into account in the economics). ------- delivered by rail in tank cars and are stored in tanks sized to provide a storage capacity of 30 days. The economic estimates consist of capital investments and both first-year and levelized annual revenue require- ments. Capital investments are based on mid-1982 costs and annual revenue requirements are projected to 1984 and are based on 5,500 hr/yr of operation at full capacity. The capital investments also include pond credits which are deter- mined by subtracting the cost of the 30- day-capacity holding pond and the 30- year-capacity waste disposal pond for the agglomeration plant from the cost of the coal-cleaning plant 30-year-capacity waste disposal pond. Process Description Oil Agglomeration with Heptane Recovery The process consists of four identical trains of agglomeration and heptane recovery equipment, supplied by a single feed tank and raw material storage and supply system. The flow diagram is shown in Figure 1. The agglomeration plant is designed to process a 10% solids slurry at a rate of 2 x 108 Ib/hr (about 4,000 gal./min of slurry) for 5,500 hr/yr for 30 years. The 10% solids slurry is recovered from the coal-cleaning plant holding pond and pumped to a scalping screen that removes particles over 0.5 mm in size. The slurry that passes through the screen is stored in an agitated tank from which it is pumped to the high-shear mix tank. Along with the slurry, 18% heptane and 2% asphalt (both based on agglomerated product from the screen) are added to the mix tank. The high degree of agitation produces very fine droplets of heptane and asphalt in which the coal and other oleophilic minerals agglomerate. The slurry flows to the low-shear mix tank which has a lesser degree of agitation and allows the small agglomerates formed in the high-shear mix tank to coalesce into particles 2 to 3 mm in diameter (an adequate size particle for thermal drying). The slurry flows by gravity from the low-shear mix tank to a vibrating screen which removes agglomerated particles. The unagglomerated particles (mostly noncoal minerals) and liquid drain to a refuse tank. The agglomerated particles are transferred by belt conveyor to the heptane recovery system. The agglomer- ation plant is designed to recover 92% of the coal in the waste and produce 289,000 ton/yr of product containing 89% coal, 8.0% mineral matter, and 3.0% sulfur with a heating value of 12,900 Btu/lb (not including residual heptane and the asphalt binder). The recovery system consists of a fluidized-bed dryer; a particle purge ves- sel; and associated gas circulating, solids collecting, and condensing equipment. The particles are transported from the feed bin in a screw conveyor and fed to the dryer through an air lock. The particles are fluidized with a heated mixture of 60% heptane and 40% water vapors which vaporizes 95% of the heptane and water in the particles. The gas from the dryer passes through a cyclone and filter to remove entrained solids. Two-thirds of the gas is recycled through a compressor and heated to 465°F before entering the dryer and one- third is passed through a water-cooled shell-and-tube heat exchanger that con- denses and cools the heptane and water to 100°F. The condensate drains to a condensate tank in which the heptane and water separate. The heptane is returned to the agglomeration system and the water is pumped to the waste pond. Approximately 97.7% of the hep- tane is recovered. Oil-Agglomeration Without Oil Recovery The nonrecovery process used No. 2 fuel oil and is based on conditions typical Asphalt Storage Tank Heptane Storage Tank Heptane Feed Tank Mud Cat Coal Refuse Pond n Low-Shear Agglomerating Tank Scalping Screen Overflow to Disposal Fluid/zed Bed Feed Dryer Bin To Drain • To Disposal Pond Steam Recycle Gas Heater _ - - Preparation Plant. fl Clean Coal Stockpile Figure 1. Oil-agglomeration process with heptane recovery. ------- of those conventional oil-agglomeration processes in which the quantity of oil used is minimized to reduce cost. The nonrecovery process differs from the recovery process by the use of a less expensive oil and a much lower oil content in the agglomerated product (6.1 % fuel oil based on the agglomerated product from the screen—see Table 1) and the absence of the oil recovery system. As a result of the lower oil content, the coal recovery is somewhat lower (90%) and the product particle size is smaller (+100 mesh). [The quantity of oil used in the agglomeration process has the greatest effect on the product particle size and coal recovery rate. The physical properties of the oil (e.g., density, viscosity) will have a slight effect, but were not evaluated in this study since their effect was considered to be very minor when compared to the effect of the oil level.] The flow diagram for the nonrecovery process is shown in Figure 2. The equipment and process descriptions for the nonrecovery process are similar to the agglomeration circuit in the recovery process except for the dele- tion of the asphalt binder for the agglom- erated coal which is not needed since the nonrecovery product is not thermally dried. Results Capital investments and first-year and levelized annual revenue requirements are developed for the base case processes just described. Several case variations and sensitivities are examined for the base cases, and the alternative of using kerosene or heptane instead of fuel oil is evaluated for the nonrecovery process. Capital Investment The summary of the capital investment estimates for the base case processes is shown in Table 2. The total capital investment for the recovery process is $9 million with the pond credit and $21 million without the pond credit. The total capital investment for the nonrecovery process using fuel oil is almost $13 million without the pond credit and a capital investment credit of $0.2 million is re- ceived with the pond credit. The major capital investment difference for the base case processes is in the total process capital. The total process capital for the nonrecovery process is approximately 43% less than the recovery process. The smaller process capital cost for the non- recovery process is due to the exclusion of the heptane recovery system, which accounts for approximately 43% of the total process capital for the recovery process. The pond credit for the nonrecovery process is approximately 6% higher than the recovery process. This is a result of the nonrecovery process using a lower oil-to-coal ratio, and thus having lower coal but higher ash recoveries, which subsequently decrease the quantity of waste solids and require a smaller refuse pond for the nonrecovery process. The pond credit has a very large effect on the capital investments and the differ- ence is illustrative of the large cost associated with pond disposal of large volumes of waste. The pond credit re- duces the total capital investments for the recovery process by approximately 57% and the nonrecovery process by more than 100%. Annual Revenue Requirements The first-year annual revenue require- ments for the recovery process are $6.4 V"" I f_ r \ Coal Refuse Pond/ Scalping Screen Overflow to Disposal °™ ggy° No. 2 Fuel Oil Storage Tank Low-Shear Agglomerating ^T^> Screen Tank High-Shear - I Mixing Tank Overflow Collection k I Tank I I Tank Agglomerated Fine Coal ci—r5> Preparation Plant Clean Coal__ Stockpile To Disposal Pond Figure 2. Oil-agglomeration nonrecovery process with fuel oil. 4 ------- Table 2. Summary of Capital Investments Investment Area Total Cost, $1000$ Nonrecovery Process Recovery Process With No. 2 Fuel Oil Total process capital Total indirect investment Working capital Other capital charges Total capital investment 10.015 6.546 909 4.029 5.668 3.704 1.109 2.279 excluding pond credit Pond credit Total capital investment 21,499 (12.168) 9.331 12.760 (12.926) 1166) million, as compared with $8.0 million for the nonrecovery process with fuel oil, including the pond credit in both cases. The difference in annual revenue re- quirements between the two processes is due mainly to the larger quantity of oil consumed by the nonrecovery process, which replaces that lost in the coal agglomerates. Even though a heat credit is applied for the oil, it is not enough to offset the difference between the cost of the two processes. The first-year unit revenue requirements are 0.86 $/108 Btu (22 $/ton) for the recovery process and 1.10 $/106 Btu (27 $/ton) for the nonrecovery process, including the pond credit in both cases. Without the pond credit, the unit costs are 1.15 $/106Btu (30 $/ton) for the recovery process and 1.42 $/106 Btu (34 $/ton) for the non- recovery process. These costs compare quite favorably with an eastern bitum- inous coal which has a heating value of 11,OOOBtu/lbandcostof1.85$/106Btu (40.70 $/ton). Case Variations Since the design data for the processes are based on laboratory tests, several case variations and sensitivities are ex- amined, as shown in Table 3, for different pond credits, raw material costs, asphalt contents, and coal recoveries for the recovery process and the nonrecovery process with fuel oil. Different heptane recoveries and purge gases are also evaluated for the recovery process, along with the effects of including a pelletizing system for the product. The effects of using alternate agglomerating agents (oils) and of using a centrifuge to separate the liquid from the coal agglomerates are also evaluated for the nonrecovery pro- cess. The pond credit has a major effect on both the capital investment and annual revenue requirements of both processes. With the pond credit, the capital invest- ments for the recovery and nonrecovery processes are 57% and over 100% lower, respectively, than capital investments without pond credits. The first-year an- nual revenue requirements for the re- covery and nonrecovery processes are increased 34% and 28% without the pond credit. The smaller percentage increase for the nonrecovery process results from the greater importance of the raw material cost, which is the predominant factor in its annual revenue requirements. The recovery and nonrecovery processes are also sensitive to raw material price, asphalt content, and coal recovery rates. The recovery process is also sensitive to heptane recovery rates and slightly sensi- tive to the type of purge gas selected. The addition of a pelletizing area to the recovery process increases the capital investments and annual revenue require- ments by 13% and 23%, respectively, and the use of a centrifuge in the nonrecovery process increases its capital investment and annual revenue requirements by 150% and 1O%, respectively. However, the cost for adding the pelletizing area to the recovery process (first-year annual revenue requirements of 1.06 $/106 Btu with the pond credit) and using a centri- fuge in the nonrecovery process (first- year annual revenue requirements of 1.21 $/106 Btu with the pond credit) is still less than the cost of the premise coal at 1.85$/108Btu. The first-year annual revenue require- ments of the nonrecovery process are projected for two other agglomerating agents (heptane and kerosene). The other nonrecovery processes in which the weight of heptane and kerosene in the product is the same as for the fuel oil process have substantially higher annual revenue requirements—$10.5 million for the kerosene process and $15.3 million for the heptane process, as compared with $8.0 million for the fuel oil process. The predominant difference is the oil costs—1.09, 1.32, and 1.60 $/gal. for fuel oil, kerosene, and heptane, respec- tively. Effect of Oil Consumption on Annual Revenue Requirements The effect of the quantity of oil used on the cost of the nonrecovery processes is shown in Figures 3 and 4 in comparison with the recovery process at different oil recovery efficiencies. The recovery pro- cess has the same oil and asphalt content in the agglomerated product as the base case processes—18% oil and 2% asphalt in the agglomerated coal. The nonrecovery process annual reve- nue requirements increase rapidly with increasing oil consumption. This is pri- marily due to the higher consumption of fuel oil which is the predominant cost factor in the first-year annual revenue requirements. The cost of the nonrecovery product with the pond credit exceeds the cost of the premise coal (1.85 $/108 Btu) at oil levels of 11 % or higher (9% without the pond credit) as shown in Figure 4. The recovery process annual revenue requirements are not related to oil content in the agglomerated coal but to the efficiency of the oil recovery system. As the recovery efficiency decreases (percent oil loss increases), the annual revenue requirements increase rapidly as shown in Figures 3 and 4. At an 18% to 20% oil loss in the recovery process, the cost of the product with the pond credit exceeds that of the premise coal at 1.85 $/108 Btu. With no pond credit, the cost of the recovery product exceeds the premise coal at an oil loss of approximately 13.5%. At the base case conditions evaluated in this study, the recovery process is less expensive to operate than the nonrecov- ery process. Also, a poorer quality of product is produced in the nonrecovery process, and there may be greater un- certainty concerning the capability of the nonrecovery process (at the low oil levels) to actually yield a product suitable for use in a pulverized-coal-fired boiler. However, the nonrecovery process could be eco- nomically competitive if lower than base ------- Table 3. Case Variation Cost Sensitivities Oil Agglomeration With Heptane Recovery Capital Investment Variation Pond credit Base case 50% of base case No pond credit Raw materials price" 80% of base case Base case 140% of base case Asphalt used* 50% of base case Base case 150% of base case Coal Recovery* 90% of base case Base case 105% of base case Heptane recovery 80% 90% 95% Base case (97.7%) 99.7% Purge gas Base case (air) Inert gas Nitrogen Pallatization Base case Base case with palletization Centrifuge Base case Base case with centrifuge Oils Base case (fuel oil) Kerosene Heptane $10" 9.3 15.4 21.5 9.29 9.3 9.4 9.1 9.3 9.6 9.5 9.3 9.1 10.0 9.6 9.4 9.3 9.2 9.3 9.2 9.8 9.3 10.5 Change. % +66 +131 -0.1 +1 -2 +3 +2 -2 +8 +3 +1 -1 -1 +5 +13 Annual Revenue Requirements* $/1OtBtu 0.86 1.00 1.15 0.79 0.86 1.01 0.78 0.86 0.94 0.95 0.86 0.82 1.96 1.34 1.03 0.86 0.74 0.86 0.74 0.75 0.86 1.06 Change, % +16 +34 -8 +17 -9 +9 +10 -5 +128 +56 +20 -14 -14 -13 +23 Oil Agglomeration With Fuel Oil Capital Annual Revenue Investment Requirements* $10* (0.2) 6.3 12.8 (0.3) (0.2) 0.1 (0.2) 1.0 1.5 (0.2) (1.2) (0.2) 0.1 (0.2) 0.1 0.5 Change. % $/10*Btu 1.10 +3,250 1.26 +6.500 1.42 -50 0.86 1.10 +150 1.60 1.10 +600 1.41 +850 1.27 1.10 -500 1.02 1.10 + 150 1.21 1.10 + 150 1.45 +350 2. 1 1 Change, % +15 +29 -22 +45 +28 +15 -7 + 10 +32 +92 *1982 dollars. "first-year annual revenue requirements in 1984 dollars. cAsphalt and heptane or fuel oil. a1%. 2%. and 3% of undried product for the recovery process; 0% and 3% for the nonrecovery process. '83%. 92%. and 97% for the recovery process; 81%. 90%. and 95% for the nonrecovery process. case oil recoveries are achieved in the recovery process. As shown in Figure 3, the first-year annual revenue require- ment with pond credits for the recovery process with only 2.3% oil loss (base case) is equivalent to the revenue re- quirements for the nonrecovey process at an oil level of 4.5% in the product, but at oil losses greater than 6.5% (recoveries less than 93.5%), the first-year annual revenue requirements for the base case nonrecovery process (oil level of 6.1 % in the product) with pond credits are less than for the recovery process. Conclusions Based on typical costs for coal, the oil- agglomeration process appears to be an ------- Recovery Process Oil Loss. % of Heptane Feed 5.0 10.0 I 15.0 I 20.0 I I a o II 10 O - H 6.0- 5.0- 4.0- £ 3.0- C 2.0- 1.0- Nonrecovery Process with No. 2 Fuel Oil Recovery Process with Heptane ~~" ~~" Premise Coal Price (1.85 S/10* Btu) • Base Case 0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 Nonrecovery Process Oil in Product, % by wt. in the Agglomerated Coal Figure 3. First-year annual revenue requirements (with pond credit) for the recovery and nonrecovery processes. 18.0 20.0 effective method for recovering coal fines from coal-cleaning plant refuse. Both the recovery and the nonrecovery processes appear to be economically feasible methods but, depending on the base case amounts of oil used in the two processes and the efficiency of the oil recovery in the recovery system, the recovery process is more cost effective. As a result of the more favorable economics for the recov- ery process at the higher oil recovery efficiencies (greater than 93%). it is recommended that this technology be tested to determine if the recovery system can be operated with the desired effi- ciency. If the recovery system cannot operate at the higher oil recovery effi- ciencies (greater than 93%), examination of the nonrecovery process may be ad- visable. Metric Conversions Readers more familiar with metric units may use the following metric conversion factors: Nonmetric Times Yields Metric Btu °F gal. in. Ib ton 1.055 5/9(°F-32) 3.785 2.54 0.454 907.2 J °C I cm kg kg ------- 5.0 Recovery Process Oil Loss, % of Heptane Feed 10.0 15.0 20.0 •6 I 6.0- 5.0- *^ jo § 4.0- § tj 5 $ .gl II „. s>a = oa 1.0- Nonrecovery Process with No. 2 Fuel Oil Recovery Process with Heptane Premise Coal Price (1.85 $/10e Btu) Base Case Figure 4. 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 Nonrecovery Process Oil in Product, % by wt. in the Agglomerated Coal First-year annual revenue requirements (without pond credit) for the recovery and nonrecovery processes. 18.0 20.0 L Larkin and J. D. Maxwell are with the Tennessee Valley Authority, Muscle Shoals, AL 35661. Julian W. Jones is the EPA Project Officer (see below). The complete report, entitled "Economic Evaluation of Oil Agglomeration for Recovery of Fine Coal Refuse." (Order No. PB 86-161 304 /AS; Cost: $11.95. subject to change) will be available only from: National Technical Information Service 5285 Port Royal Road Springfield. VA 22161 Telephone: 703-487-4650 The EPA Project Officer can be contacted at: Air and Energy Engineering Research Laboratory U.S. Environmental Protection Agency Research Triangle Park. NC 27711 U. S. GOVERNMENT PRINTING OFFICE: 1986/646-116/20802 ------- p- a> co oo < en a> O S w c IS) V) CD 10 q = D - 3 a B) C s. c O: I 0) 00 ------- |