Economic factors affecting coal preparation: plant design worldwide and case studies illustrating economic impact
Abstract:
The economic drivers behind the differences in design of plants treating coal destined for thermal and metallurgical coal markets worldwide are considered. Differences between plant designs in Australia, South Africa and the United States for both coal types will be discussed. Environmental constraints on plant design will be reviewed, in particular the means of dealing with ‘dry rejects disposal’ and water scarcity. Dealing with the ever-deteriorating quality of plant feed material, with the inevitable increase in near density material, will also be considered.
coal cleaning circuit optimization
coal processing equipment selection
coarse coal only cleaning plants
14.1 Introduction
This chapter begins with a consideration of the economic drivers behind the differences in design of plants treating coal destined for thermal and metallurgical coal markets worldwide. Further, the differences between plant designs in Australia, South Africa, and the United States for both coal types will be discussed. Environmental constraints on plant design will be reviewed, in particular the means of dealing with ‘dry rejects disposal’ and water scarcity. Dealing with the ever-deteriorating quality of plant feed material, with the inevitable increase in near density material, will also be considered.
14.2 Current steam coal circuit design
Most thermal coal sales contracts are driven by the desired calorific value (CV) of the coal delivered to the customer. Typically, minimum CV levels will be specified, with maximum tolerated ash, moisture, and sulfur levels, together with limits for other parameters such as Hardgrove Grindability (degree of difficulty in grinding) and ash fusion temperature (indicator of slagging risk in a boiler). Plant design therefore should be centered on delivering the desired CV coal at the maximum plant yield while satisfying the moisture ash and sulfur constraints and keeping other parameters in mind.
Fines dewatering can be the controlling factor in the acceptability of the final product. Conventional dewatering technology for − 0.15 mm coal fraction usually consists of centrifugation in screen-bowl centrifuges, or vacuum filtration. These techniques produce incremental moistures in the ultrafine particle sizes range of 50%. Typical dense medium (DM) plants when operating in the 1.6–1.7 relative density (RD) range generate incremental ash contents in the mid 30 percentiles, with associated moisture contents in the ± 5% range. Assuming all other constraints are met, the economics therefore drives the ultrafines to be discarded and plants to be run at higher DM circuit relative densities.
Effectively, over 55% of inert material (ash + moisture) is discarded by not recovering the minus 0.045 mm material and replacing it with approximately 40% inerts contained in the middlings material. This is because the higher energy content of the middling’s material allows more of this to be blended back than the material lost by discarding the ultrafines.
This phenomenon has driven thermal coal treatment circuit design in Australia, the USA, and South Africa, which commonly all discard nominal − 0.15 mm material. A typical Australian steam coal preparation plant is shown as Fig. 14.1 (Firth et al., 2006). An American thermal coal plant would be similar except that instead of two circuits (DMC & spirals) three circuits will be used (DM bath, DM cyclone, and spirals). More recently, as discussed below, deslime column flotation has been added, as shown in Fig. 14.2 (Bethell and Barbee, 2007).
US coals tend to be more readily liberated than Australian coals. By maintaining a coarse plant feed top size of 150–200 mm, particle surface area is minimized. Therefore, plant product moisture content is minimized. This allows maximization of plant DM operating RD, which in turn maximizes yield.
In Australia, it has been found that crushing plant feed to minus 50 mm gives better coal liberation. The yield enhancement from this characteristic more than offsets the negative impact of higher plant moisture content.
Historically, South Africa has had a circuit design similar to that of the USA, although equipment selection was different (DM drums vs DM baths). Also, the South Africans tend to produce two products: a high quality export thermal product, and a high-ash domestic steam coal. However, recent circuits designs (Fig. 14.3), most notably at the Phola and Mafube mines, have crushed the plant feed to a 50 mm top size for liberation reasons. Both plants, however, use coarse and fine DM cycle circuits to achieve enhanced recovery of the finer DMC feed size fractions (Cresswell, 2010).
Recently in the USA, incorporation of deslime column flotation circuits has gained great popularity and typically provide excellent financial return. An example of one such installation is the Lone Mountain column project in Virginia (Baumgarth et al., 2005).
Testing of the flotation characteristics of the Lone Mountain plant feed coals showed excellent coal recovery at low ash. Frothing the − 0.15 mm material in columns was projected to generate a high-moisture product. This high moisture level would inhibit the capability of back blending higher ash raw coal fines into the final product, preventing − 0.15 mm frothing from being economical.
However, flotation release analysis testing on a simulated deslimed feed also showed excellent recovery at low ash. Past experience in handling this type of material (0.15 × 0.045 mm) in screen-bowl centrifuges with spiral product have indicated that low incremental moistures could be achieved on the nominal column product. This product would not interfere with the ability to back blend higher ash raw coal fines. The column product would be totally accretive. Based on these projections, a single 4 M diameter Eriez Coalpro column was installed at Lone Mountain in September 2004, on a nominal 0.15 × 0.045 mm feed.
The column unit has consistently produced 20 + t.p.h. of concentrate. The Decanter screen-bowl centrifuge that has replaced an EB42 screen scroll dryer (previously treating spiral product) has shown a decrease in product moisture. The total plant product moisture also has decreased, thereby allowing the back blending of additional higher ash raw fines.
In these deslime column circuits, the − 0.15 mm material, which would normally be discarded, is instead treated in small diameter hydro-cyclones (150–165 mm diameter), which remove ± 70% of the slimes (− 0.045 mm) material. The cyclone underflow is fed to column flotation, where this highly selective process generates a low ash, relatively low incremental moisture product.
The incremental inerts on this nominal 0.15 × 0.0045 mm product are typically below 25%, making it a highly desirable product. In many cases, including the above-mentioned Lone Mountain example, plants retrofitted with the deslime column circuit replace originally installed fine coal centrifuges with screen-bowl centrifuges on combined spiral/deslime column concentrate and produce a reduction in overall plant moisture content. This is achieved because the increased incremental moisture of the column concentrate is overcome by the reduction in moisture content of the 1 × 0.15 mm material. A schematic diagram of the deslime column flotation fine coal circuit is shown in Fig. 14.4.
14.3 Current metallurgical coal circuit design
Metallurgical coal sales contracts typically specify expected coking characteristics of the delivered coal including reflectance, fluidity, dilatation, and numerous other coking characteristics. Ash, phosphorus, and sulfur levels, which are also critical in coking coals, are always expressed on a dry basis, with specified targets and maxima. Coal is purchased on an as-received basis, moisture contents usually being contracted on this basis also.
Plant design therefore targets achieving desired as-received (AR) moisture content and dry ash, phosphorus, sulfur, and other contaminant levels commensurate with acceptable coking characteristics.
For several coking and producing regions only certain size fractions possess acceptable coking characteristics and, consequently, part of the plant product (typically the coarsest fractions) is sold as steam coal, for example the Rangal coal measures in Australia.
Major circuit design differences are once again seen between US and Australian metallurgical coal plant design.
Australian coking plants historically have had two circuits: DM cyclones on 50 × 0.5 mm material, and flotation on the − 0.5 mm.
Traditionally conventional froth flotation equipment was utilized. However, after the successful installation of column flotation at Middlefork in the USA in 1990, most subsequent Australian plants have used column flotation, mainly the Microcell or the Jameson types.
Column flotation provides superior selectivity compared to conventional flotation, particularly for coals rich in clays. Conventional cells carry entrained slimes into the product, which are minimized by the deep froth depth and wash water used in column flotation.
In the two-circuit met. Plants (DMC /froth flotation), inevitably some misplaced oversized material enters the flotation circuits due to worn deslime screen panels, particle shape effects, etc. Unfortunately, flotation kinetics are poor on + 0.3 mm material and much of the + 0.5 mm reaching the froth cells is lost. Typically, Australian coking coal plants need to run fairly low RD (− 1.5) to achieve desired ash quality specifications. The inability of spirals to separate in this range has precluded them, to a large part, from coking coal plants. Recent innovations in hindered-bed separation have allowed an intermediate circuit to come into favor for Australian coking coal plants. In particular, the new reflux classifier, capable of high efficiencies over a fairly broad size range and a fairly wide RD of separation range, has emerged for coking coal plants, as well as for steaming coal plants where appropriate (Galvin et al., 2010).
The hindered-bed devices, which are capable of lower cut points than the spirals, allow for more efficient recovery of the 1 mm × 0.25 mm material than the traditional conventional froth circuits. A schematic drawing of the Reflux Classifier is shown in Fig. 14.5. The coarse oversize material previously lost in the flotation circuits can now be captured in the hindered settling separator. Also, by increasing the aperture of the desliming screens from the traditional 0.5 m–1 mm, greater screen capacity and therefore increasing plant capacity are achieved.
Filtration, either by vacuum disk, drum, or horizontal belt filtration is practiced on the froth concentrate to maximize fines recovery. Customers of Australian metallurgical coals are more tolerant in their moisture specifications into their export-dominated markets than USA domestic or export contracts. The higher shipped moisture content tolerance for Australian coals enables superior fines recovery to occur.
US metallurgical coal plants traditionally employed a combination of conventional froth flotation on the − 0.5 mm fraction with thermal drying to meet the more stringent US coking coal sales moisture specifications. The operating and maintenance (O&M) cost, and the difficulty of obtaining and maintaining permits for thermal dryers, have seen their gradual demise in the USA, although many of the Canadian coking plants employ thermal drying including several newer facilities (Hogg, 2007).
US metallurgical coal plants have evolved to either DM bath with DM cyclone or DM cyclone alone for + 1 mm material with compound spirals to treat minimal 1 mm × 0.2 mm material. Good liberation has allowed the high cut-point spirals to be a highly efficient part of modern metallurgical coal preparation plants. The benefit of the compound spiral vs a conventional spiral can be seen in Fig. 14.6 (Luttrell et al., 2003). Depending on the level of clay content, conventional or column flotation are used, either in a ‘by-zero’ or deslimed circuit.
Due to their high-moisture product content, filtration is not widely practiced, unless thermal drying is available, in newer US coking coal plants.
The fine coal fraction tends to be dewatered in screen-bowl centrifuges being fed with combined spiral and froth concentrate (Fig. 14.7).
Unfortunately, the screen-bowl centrifuges, although very efficient for + 0.045 mm material, will often capture only about 50% of the − 0.045 mm material, leading to coal losses. A recently developed high speed centrifuge the ‘Centribaric’, has shown great promise in recovering ultrafines lost in normal screen-bowl centrifuges, and may well become an integral part of future circuits (Shultz et al., 2010).
Alternatively, the deslime column circuit mentioned in the thermal coal section is used. These deslime circuits have some losses of coal in the ultrafines, but moisture specs are met and + 0.045 mm recovery is maximized.
In both the USA and Australia the trend is to use fewer units of larger equipment, in both thermal and metallurgical coal facilities. Single unit operation circuits are favored for their simplicity, low capital, operation and maintenance costs. Major processing advantages are also seen with elimination of the distribution problems associated with multi-unit circuits.
14.4 Case studies
Case studies were conducted to evaluate the impacts of the different deslime and flotation circuit configurations on overall economic performance. Two series of studies were performed, one for the thermal market and one for the metallurgical (coking) market (Bethell and Luttrell, 2005).
14.4.1 Metallurgical coal market
To illustrate the impacts of desliming on plant performance, a case study was performed in which different circuit configurations were evaluated for a plant producing metallurgical coal with contractual limits of 5.7% ash (dry basis) and 7% moisture. As shown in Table 14.1, three circuit configurations were considered for conventional flotation (Cases A1–A3) and for column flotation (Cases B1–B3). These configurations included (i) non-deslimed flotation with vacuum filters, (ii) non-deslimed flotation with screen-bowls, and (iii) deslimed flotation with screen-bowls. Deslimed flotation with vacuum filters was not considered a viable option, since the decision to discard the ultrafine coal in the desliming cyclones made the high cost of installing and operating filters impossible to justify.
Table 14.1
Case study comparing different circuits for the metallurgical coal market (contract specifications: < 7 % moisture, < 5.7% ash {dry})
For conventional flotation and vacuum filters (Case A1), an acceptable ash could be made only by dropping the coarse circuit cut points to 1.50 RD. The total moisture content produced by this combination was 8.4%, which greatly exceeded the contract moisture limit of 7%. As such, filtration is a viable option for this configuration only when thermal drying is used downstream, or where moisture specs are lower (Australia). Conventional cells with screen-bowls (Case A2) lowered the total moisture to 7.4% and allowed the coarse circuit cut point to be raised to 1.60 RD before the ash limit was reached. However, this option was also not acceptable, since the 7.4% moisture content still exceeded the contract specification. The only configuration with conventional cells that could meet the target moisture was the deslime circuit with screen-bowls (Case A3). This circuit, which provided very low product moistures, generated 1013 t/h of saleable coal at the moisture limit of 7%. As a result, the deslime option is attractive for metallurgical (coking) coals when moisture limits cannot be met.
The circuits incorporating the column flotation technology are shown in the bottom half of Table 14.1 as Cases B1–B3. In general, column cells are an effective means of removing fine clay slimes from the flotation concentrate. As a result, the non-deslime column circuits (Cases B1 and B2) provided substantially higher tonnages of clean metallurgical coal than did the conventional flotation circuits (Case A1 and A2). Column cells with vacuum filters (Case B1) produced the highest tonnage (1047 t/h) of clean coal without exceeding the contract ash limit, but the 7.7% moisture content once again exceeded the moisture specification. This would probably be acceptable in Australian markets, and would make this circuit option the most attractive in that circumstance. On the other hand, the use of screen-bowls in place of the vacuum filters (Case B2) reduced the moisture to a value of 7.1%, which may be acceptable since it is very close to contract specification. This configuration produced 13 t/h more coal (i.e., 1026 vs 1013 t/h) than the deslime circuit with conventional flotation and screen-bowls (Case A3). This tonnage gain represents an increase in revenue of about $7.8 million annually (i.e., 13 t/h × 6000 h/year × $100/t = $7.8 MM/year). In this case, the column technology made it possible to recover valuable ultrafine coal that would otherwise be discarded by circuits that employ desliming cyclones ahead of flotation. Unfortunately, the column cell and screen-bowl circuit may not be acceptable for contracts with tighter moisture limits. This problem can also be overcome by using deslime cyclones ahead of the column cells and screen-bowls (Case B3), but this configuration provides only a slightly higher tonnage (1014 vs 1013 t/h) of clean coal than the otherwise identical conventional flotation circuit. Thus, the use of columns may be difficult to justify when deslime cyclones are used together with screen-bowls to provide products with low moisture contents. On the other hand, the circuit incorporating column cells and vacuum filters (Case B1) would significantly outperform all other circuits in terms of coal production (967 t/h of dry clean coal) if no moisture constraint existed and/or thermal dryers were available.
14.4.2 Steam coal market
Unlike metallurgical coal contracts, as mentioned earlier steam coals are most often valued on the basis of delivered heat content. Although maximum ash and moisture levels are specified (AR basis), meeting minimum CVs is normally the major hurdle to plant yield maximization. This key difference in contract specifications makes ash and moisture virtually interchangeable provided that neither is contractually exceeded. For example, a coal with a moisture-ash-free (maf) heat value of 8333 kcal/kg can meet a 7000 kcal/kg contract specification equally well by shipping an 8% moisture and 8% ash product or by shipping a 6% moisture and 10% ash product (all on an AR basis). The major advantage of minimizing moisture is that for every ton of water eliminated, typically 2 + tons of middlings containing 35–45% ash can be added back to the product by raising cut-point gravities in the coarser circuits. The replacement of inert matter (ash and moisture), having no heating value, with carbonaceous middlings provides a strong financial incentive for using desliming cyclones in plants producing steam coals.
Table 14.2 summarizes data for a case study conducted for a typical plant producing steam coal. The sales contract required < 8% moisture, < 14% ash and a heating value of > 6690 kcal/kg. Three scenarios were considered: (i) discard the − 0.15 mm (100 mesh) fines without treatment, (ii) recover the fines using a conventional flotation bank without desliming, and (iii) recover the fines using a column circuit after desliming. The results show that the plant produced 899 t/h of clean coal at the target heat content of 6690 kcal/ kg when no flotation was used and the fines were discarded (Case A). When using conventional flotation and filters without desliming (Case B1), an acceptable heating value could be achieved, but only after rejecting 29.7% ash middlings by lowering the cut point in the coarser circuits from 1.68 to 1.50 RD. Unfortunately, this option increased the moisture content of the plant to 8% and lowered the overall tonnage to 882 t/h. On the other hand, the use of deslime cyclones, flotation, and screen-bowls (Case B2) substantially reduced the moisture and allowed the 1.50 × 1.65 RD middlings to be recovered without exceeding the 6690 kcal/kg contract limit. The desliming step increased the net tonnage to 923 t/h, a net gain of 24 t/h over the base scenario (Case A). This represents nearly $6.5 million annually of additional revenues (24 t/h × 6000 h/year × $45/ton = $6.48 MM/year). Obviously, desliming to reject high-ash, high-moisture material is very attractive for this particular coal.
Table 14.2
Case study comparing different circuits for the steam coal market (< 8% moisture, < 14% ash, 6690 kcal/kg)
Table 14.2 also shows the results obtained using column flotation cells with filters (Case C1), column flotation cells with screen-bowls (Case C2), and deslimed column flotation with screen-bowls (Case C3). When using vacuum filters, the column cells reduced the ash content of the froth product to 6.1% (Case C1), compared to 8.9% for conventional flotation (Case B1). This reduction allowed the 1.50 × 1.58 RD middlings to be recovered, so that 34 t/h (916 – 882 = 34 t/h) more production could be realized compared to the conventional circuit. The use of screen-bowls with the columns (Case C2) only slightly increased the clean tonnage by making it possible to pull harder during flotation. On the other hand, the deslime column circuit with screen-bowls (Case C3) made it possible to send both the froth product and the middlings (1.50 × 1.68 RD) to clean coal without exceeding either the ash, moisture, or heating value limits stated in the contract. This configuration made it possible to increase the tonnage of saleable coal to 929 t/h. While the circuit provided just 6 t/h more production than the equivalent conventional circuit, the productivity gain is worth more than $1.6 million annually (i.e., 6 t/h × 6000 h/year × $45/t = $1.62 MM/year). In many cases, the higher revenue is sufficient to justify the marginally higher capital and operating costs for the column cells. More importantly, this analysis shows that the use of deslime cyclones ahead of flotation provides tremendous financial advantages over non-deslimed circuits for both the column and conventional circuits.
The results obtained for the case studies discussed above are fairly typical. Field data suggest that conventional flotation cells typically generate products having − 45 μm solids with incremental moistures in the 50% range. Entrained clay slimes can have 80% ash on a dry basis, which equates to 90% inerts passing into the froth product on an AR basis. Even column circuits treating − 0.15 mm (100 mesh) material with low − 45 μm ash levels will typically have incremental inerts in the range of 55% (i.e., 50% moisture and 5% ash AR). Therefore, the recovery of ultrafines is generally uneconomical for the steam coal market. This generalization is particularly true in the Appalachian coalfields in the USA, where steam plants treat coals with high clay contents and operate heavy medium circuits in the 1.50–1.60 RD range. Most plants do not have thermal drying; consequently maximum plant yield and the best financial returns can be obtained by designing a deslime column flotation circuit with screen-bowl centrifuge dewatering. Once the plant RDs have been maximized (+ 1.70 RD) and room still exists for higher moisture fines, then obviously a column concentrate having − 45 μm product at 55% inerts can be recovered, although its value will be low.
It should also be noted that the capital costs will be much higher for non-deslime fines circuits. For a typical plant, capital costs for a by-zero (non-deslimed) circuit can be expected to be at least double the costs of the deslime circuit. The by-zero flotation cells have to handle approximately four times the volume that the deslime circuit would process. Also, carrying capacity, normally the major driver for circuit size and number of required column cells, is a function of particle surface area. The carrying capacity will often be half the value for by-zero applications in comparison to deslime scenarios, thus requiring twice the column capacity to get the same clean coal yield.
Where the coal seam does not contain significant amounts of ultrafine clay, conventional flotation may be competitive and the benefits of column flotation minimized. Also, the presence of existing thermal drying capacity makes a by-zero circuit much more attractive, since thermal drying overcomes the very high incremental moisture associated with ultrafine solids. If the coals are relatively free of clay, and the plant RD cut points are already set to maximum practical values, conventional flotation may also be economically justified as well.
14.5 Poorly liberated coal
Throughout the coal-producing regions of the world, the best, most liberated coals have already been mined. This leaves coal preparation with major challenges in dealing with coal having more near-gravity material. Many of these coals, in India, Mozambique and South Africa, will require substantial crushing to provide material liberated enough to generate a marketable product at economical yields. Crushing down to a top size of about 12 mm, and even finer, will probably become commonplace. Some deshaling, i.e., removing rejects prior to normal treatment, may be appropriate ahead of the comminution stages to reject high-ash material (Zaija et al., 2007).
Faced with a very finely sized plant feed, major challenges will be encountered in the areas of efficient fine coal cleaning and dewatering.
Conventional DM cyclone circuits will almost certainly be used to treat the coarser material, although bottom size feeding these DMC units may revert to the traditional 0.5 mm compared to the coarser cuts seen more recently. This, coupled with the use of smaller diameter DMC units, will allow the efficient treatment of + 0.5 mm particles to be maximized.
As far as the − 0.5 mm (or − 1 mm) is concerned, Reflux Classifier circuits to treat down to 0.25 or 0.15 mm will be applicable for low density cuts and compound spirals for higher density separations. The reintroduction of fine DMC cycloning, which is being pioneered in South Africa (Dekorte, 2002), may well also provide critical and highly efficient cleaning in the 0.5 × 0.15 mm range (Fig. 14.8). The bottom size feeding these units will be determined by the froth flotation characteristics of the ultrafines. Good ‘floating’ coals will probably be treated in flotation circuits having a top size of 0.25 mm. For poor floating coals, fine coal DMCs, spirals, or reflux classifiers will probably be fed material down to 0.15 mm with discard of the ultrafines.
Screen scroll dryers will most likely be used on the DMC product and screen-bowl centrifuges on the spiral/Reflux/fine DMC product. Either vacuum filtration or pressure filtration (depending on product moisture content and market moisture constraints) will be used to dewater the froth concentrates. Examples of vacuum and pressure filtration are shown in Figs 14.9 and 14.10.
With the ever-increasing fines loading, high efficiency screening at fine sizes will also be critical. Efficient fine screens, such as the Derrick Stack sizer (Fig. 14.11) (Brodzik, 2007), will no doubt find considerable use.
The use of online nuclear analyzers will become ever more important as the quality of coals mined deteriorates and exacting quality specifications must be met from inferior coals run at low operating RDs. Similarly, plant control circuits will need to be maintained at a very high level in dealing with coals possessing high percentages of near-gravity material.
Some form of briquetting or agglomeration may be necessary in view of the large percentage of fines in the finished product.
14.6 Water constrained plants
Many areas of the coal-producing world have severe shortages of water. Coal-processing plant design will need to be innovative and resourceful in conserving their water supplies to be viable.
Several alternative strategies are available to deal with dearth of water. These include closed water circuits for conventional plant circuit design, ‘coarse wash only plants,’ and dry cleaning circuitry. Also, desalination plants are being introduced, significantly pushing up treatment costs.
14.6.1 Closed water conventional circuitry
Whenever possible, current processing plants dispose of fine waste material in slurry form. This is done by pumping the plant tailings, predominantly thickener underflow, to dams, impoundments, slurry cells, underground, etc. Typically, where fresh water supplies are abundant, these methods provide the most economical disposal methods.
With scarcity of water supply and environmental constraints in many regions of the world, this is no longer possible. Alternatives include the use of belt presses, and plate and frame filters, which allow fine refuse to be disposed of ‘dry,’ providing good husbanding of the scarce water resources. Many plants in the USA, Australia, China, and South Africa have already moved into this mode of fine refuse disposal. A further alternative to minimize water consumption and provide a more handleable product is the ‘deep cone thickener.’
This device is widely used in the diamond industry to thicken tailings into a ‘stackable paste’ for disposal. This technology has been applied in the coal industry at Lone Mountain plant in the USA (Fig. 14.12) (Gupta, 2010). This plant now pumps conventional thickener underflow to a deep cone thickener, which replaced four belt presses. The ultimate product of fine refuse (~ 100 tonnes/h) is disposed of at about 45–50% solids at the deep cone thickener feed rate of 900 m3/h, delivering this material to the storage area. This installation has allowed substantial flocculant savings to occur and maximizes the return of clarified water to the plant.
14.6.2 Coarse wash only plants
Partial washing of coal, with back blending of raw fines with the washed coarse coal, in some circumstances can provide a market-acceptable product quality. This type of plant is designed for minimum water consumption, and fine refuse disposal. Typically, these plants will be used when the quality of the raw coal material, which traditionally served thermal coal markets unwashed, has deteriorated to the point that its quality raw is now unacceptable to the customer. Examples of this phenomenon include the Western USA, where coal seams mined in Colorado and Utah have become progressively thinner and higher in ash, delivering run-of-mine (ROM) products no longer saleable on a raw basis.
In these cases, and in similar circumstances in some South African mines, washing of the coarse coal only will provide sufficient beneficiation to render an unsaleable ROM coal saleable. Typically in these plants ROM coal is screened at between 6 and 9 mm. The oversize material passes to a DM bath with the fines by-passed to product.
Conventional screening techniques will not provide efficient separation dry at fine sizes. Consequently, innovative screen technology such as roller screens and ‘flip flop’ screens have been utilized to achieve high levels of screening efficiency.
Recent plants incorporating this circuit design concept include Phoenix in South Africa and the Castle Valley (Roxon Roller Screen) and West Elk (Bivitec screen) plants of Arch Coal in the USA.
In both of the latter plants, dry screening is achieved on the novel screen technology with the oversize (± 6–8 mm) treated in a DM bath circuit at high efficiency levels. The Castle Valley Circuit is shown as Fig. 14.13 (Kelley and Bethell, 2008). Figure 14.14 shows the West Elk plant. The flow in the plant from the Bivitec screens to the pre-wet screens and thence to the DM bath followed by appropriate screening, crushing and drying can clearly be seen.
Wet fines generation and fine refuse disposal needs are minimized, and therefore water requirements are also minimized, which is essential in the dry, environmentally constrained Western USA.
The economic benefits of the coarse ‘wash only’ plants vs conventional washing on the ‘marginal direct ship’ coals are enormous, both in terms of reduced plant capital cost and increased plant yield. However, it has to be borne in mind that this particular plant design has limited application. For the Western USA coals treated in ‘coarse wash only’ plants, an advantageous feature has proved to be the much higher ash level in the coarse material than in the fines. This, coupled with typically extremely well-liberated coarse material, has enabled these plants to produce a very saleable combined product of washed coarse coal and by-passed fines.
14.6.3 Dry cleaning
The ability to have a dry cleaning process for coals, particularly in arid regions, or in those lacking suitable water supplies or wet waste disposal sites, is obvious.
Air Jigs and Air Tables, many resurrected from old designs, have been developed to fulfill this need. Although the levels of processing efficiency reached by these dry cleaning devices are considerably less than for DM processes, the ability to clean coal with no water requirements has major benefits. To date, cleaning in the 50 mm × 6 mm size range has proved successful, particularly where the devices have been used as a destoner (deshaler) reducing ash level of previously unsalable coal into a marketable range.
The FGX device (Fig. 14.15) developed in China has found widespread worldwide use in deshaling (Orhan et al., 2010). Work continues to drive the efficient cleaning size for these units finer to increase their usage and enable dirtier coals to be handled. For good dry cleaning to occur it is important that the feed material be reasonably dry. These units operate similarly to the wet shaking tables of yore, the shaking action being supplemented by fluidizing air.
14.7 Future trends
From the foregoing, it is apparent that coal processing engineers in the future will be facing major challenges. They will need to produce a high quality product from inferior, middlings-rich coals, constrained by even scarcer water sources and stricter environmental regulations.
The need to continue to improve fine wet-cleaning techniques, such as spiraling and hindered-bed separation, as well as successfully driving down the bottom size treated by DMCs, will be paramount. Continual improvement in techniques to effectively dewater the fines coal products from the new plants will be required. Depending on handleability and dustiness, the briquetting process may well be required to produce a similar product.
Improvements in dry screening efficiency and processing will be extremely helpful in arid areas.
The area of ultrafine wet coal cleaning currently reserved for flotation (column and conventional) will need to be expanded to include efficient density separation to process oxidized coals, or those not efficiently treated by flotation (highly pyritic coals), and successful applications of devices such as the Falcon Concentrator and Kelsey Jig modified to achieve efficient separation in coal cleaning may well serve this purpose.
Through innovation and fortitude the coal processing engineers of the future will rise to the challenges facing them, as this group has always done in the past. This will be driven by a ‘coal hungry’ global market.
14.8 References
Baumgarth, T., Gupta, B., Bethell, P., Recovering an additional 20 tph of coal through a deslime column flotation circuit addition at Lone Mountain Processing. Proceedings Coal Agg Prep, 2005:41.
Bethell, P., Barbee, C., Today’s Coal Preparation Plant: a Global Perspective. Designing the Coal Preparation Plant of the Future Proceedings SME, 2007:9.
Bethell, P., Luttrell, G., Effects of Ultrafine Desliming on Coal Flotation Circuits. Proceedings Century of Flotation Symposium Brisbane, 2005:43.
Brodzik, P., Application of Derrick Corporations stack sizer in clean coal spiral product circuits. Designing the Coal Preparation Plant of the Future Proceedings SME, 2007:89.
Cresswell, G., Process Design of the Phola Coal Preparation Plant Proceedings ICPC Lexington, 2010:66.
Dekorte, D.J., Dense media benefication of fire coal revisited ICPPC Johannesburg, 2002.
Firth, B., Luttrell, G., Bethell, P., Comparison of Preparation Plant Design for Metallurgical and Steam Coal Applications Proceedings ICPC Beijing, 2006:98. [2–5].
Galvin, K., Walton, K., Zhou, J., Gravity Separation and Classification of Fine Coal Using the Hydrodynamics of Inclined Channels. Thirteenth Australian Coal Prep Conference, 2010:224.
Gupta, B., Deep Cone Thickening at the Lone Mountain Plant Proceedings ICPC Lexington, 2010:674.
Hogg, J., A Canadian View Proceedings Coal Prep, 2007:199.
Kelley, M., Bethell, P., The Design Commissioning and Operation of the Castle Valley Plant 2008 SME Meeting Salt Lake City, February 2008.
Luttrell, G., Stanley, F., Honaker, R., Bethell, P., Operating Guidelines for Coal Spirals Proceedings Coal Prep, 2003:69.
Orhan, E., Orgun, L., Bakialtiparmak, Application of the FGX Separator in the Enrichment of Coal Proceedings ICPC Lexington, 2010:562.
Shultz, W., Keles, S., Luttrell, G., Yoon, R.-H., Estes, T., Bethell, P., Development of the Centribaric Dewatering Technology Proceedings ICPC Lexington, 2010:488.
Ziaja, D., Yannoulis, G., Is there anything new in coarse or intermediate coal cleaning? Designing the Coal Preparation Plant of the Future Proceedings SME, 2007:43.