United States Environmental Protection Agency Industrial Environmental Research Laboratory Research Triangle Park NC 27711 Research and Development EPA-600/S7-83-040 Sept. 1983 4>EPA Project Summary Control of Criteria and Non- Criteria Pollutants from Coal/Oil Mixture Combustion M. E. Kelly, R. M. Parks, and J. H. E. Stalling As the availability and cost of oil have become uncertain in recent years, the need for the U.S. to reduce its depend- ence on oil has prompted significant efforts by government and industry in finding alternate fuel sources. This posture has been strengthened by the Powerplant and Industrial Fuel Use Act of 1978 that prohibits the use of gas and oil in new boilers without special exemption. The Department of Energy (DOE) has taken an active role in developing one such alternate fuel technology: coal/oil mixture (COM) combustion. Recognizing that environmental considerations must be made in any evaluation of fuel conversion to COM, DOE and EPA contracted with Radian to identify and assess the effectiveness of currently available methods of controlling the release of criteria and non-criteria (trace elements) pollutants from the combustion of COMs. The report presents this assessment and compares the costs and effectiveness of various control technologies found to be applicable to emissions from boilers firing COM. Emissions from COM combustion were characterized using data from various tests. The pollutants examined most closely were particulate matter, SO2, and NO*. Trace element emissions and emissions of polynuclear organic material were also examined. Conven- tional emission control techniques were determined to be the most effective in reducing emissions from COM com- bustion. Emission rates and associated costs for emissions control of particulate matter and SOz were assessed for four different COM compositions and various boiler sizes. Fabric filters and electro- static precipitators were considered for particulate matter control; wet flue gas desulfurization (dual alkali), dry scrub- bing, and dry injection techniques were considered for three levels of SO2 control (50%, 70%, 90%). Low-sulfur fuels were also examined as an addi- tional alternative. This Project Summary was developed by EPA'3 Industrial Environmental Research Laboratory. Research Tri- angle Park. NC. to announce key find- ings of the research project that is fully documented in a separate report of the same title (see Project Report ordering information at back). Introduction The need for the U.S. to reduce its dependence on oil and expand its use of coal has prompted DOE to take an active role in developing Coal/Oil Mixture (COM) combustion technology. Sufficient technical and economic information has been developed to allow private sector implementation of COM combustion, and a number of industrial and utility energy producers are assessing the use of COM at their facilities. As COM combustion technology be- comes more widespread, DOE and EPA recognize that certain environmental considerations will have to be addressed in order to expand the use of COM in an environmentally acceptable way. Specif- ically, it will be necessary to determine the potential emissions from sources burning COMs and evaluate, assess, and compare the effectiveness of control ------- technologies in controlling the criteria and non-criteria pollutant emissions. Radian has been under contract to EPA and DOE to identify and assess the effectiveness of currently available methods of controlling the release of criteria and non-criteria (trace elements) pollutants from the combustion of COMs. The report of this study contains several chapters that focus on the various considerations and assessments neces- sary in making a conversion from oil or gas to COM. Chapter 2 presents some of the technical and economic factors that determine the suitability of COM as a fuel and identifies boilers that might be candidates for COM conversion. Chapter 3 discusses the emissions that have been measured to date on combus- tion units burning COM. The criteria pollutant emissions that are discussed are particulate matter, SO& and NO>. Non-criteria pollutant emissions discus- sed are the selected trace elements. As, Be, Cd, Cr, Ni, Se, V, and Hg, as well as polynuclear organic matter (POM) emis- sions. Chapter 3 also briefly discusses control technologies applicable to boilers that could be converted to burn COMs. Several of these control devices are selected as most promising for further technical and economic study in the remainder of the report. Chapter 4 presents the environmental impacts of the selected control techno- logies, including the primary pollutant control capability and any secondary environmental impacts. The potential for multipollutant control is also discussed. Chapter 5 addresses the cost impacts of the technologies discussed in Chapter 4 in terms of capital costs, annual operation and maintenance costs, annualized costs, and cost effectiveness. Compar- isons of multipollutant controls, add-on controls, and low-sulfur COMs are also presented in Chapter 5. Use of COM in Existing Boilers COMs typically consist of 20-50% by weight coal dispersed in oil or in an oil/water mixture (known as COW— coal/oil/water). The coal is usually ground to at least 70% minus 200 mesh. A No. 6 heavy residual is typically chosen over lighter distillate oils for COMs. In addition to being less expensive than distillate oil, heavy residuals reduce coal settling problems with COM, relative to lighter oils. Water is sometimes added to the COM as a stabilizer; most commonly the added water is 5-20% by weight. The suitability of converting a boiler currently firing oil or gas to COM depends on several technical and economic factors. Whether a specific boiler is a likely candidate for COM depends on the age of the boiler (remaining useful life), boiler design, size and capacity factor, geographical location (proximity to COM fuel supply), site-specific boiler modifica- tions required, existing emission control equipment, if any, and emission controls required by the applicable environmental regulations. Technical Considerations Three factors must be evaluated in determining the technical suitability of converting existing oil- or gas-fired boilers to COM: COM fuel properties, COM combustion characteristics, and boiler modifications required to accom- modate COM-firing. Other than combustion characteristics, the most important COM fuel properties to be considered are viscosity, stability, and abrasiveness. The viscosity and stability of the COM are interrelated: generally, the more viscous the COM is the less tendency there is for fuel instability (settling of coal particles). Chemical stabilizing additives (typically emulsifying agents, gelling agents, or surfactants) have been developed to improve COM stability by keeping the coal particles suspended in the oil. Although maximum coal concentration is desirable for economy and stability, there appears to be an upper limit of about 60% coal in the COM. Although 60% COM may be possible, COM is typically available with 40-55% coal in oil. The increased abrasiveness of COM relative to oil can potentially cause erosion in pipe bends, pumps, valves, and burners. These potential problems can be minimized by proper material selection, reduced fuel velocities, and more finely ground coal. The boiler modifications required in converting from oil or gas to COM are quite site-specific, and depend on such boiler design factors as tube spacing, burner design, furnace size, and bottom ash removal capability. Boilers originally designed for oil- or gas-firing commonly have narrower tube spacings than boilers originally designed for coal -f iri ng, maki ng coal-fired boilers more ideally suited, in that respect, for conversion to COM combustion. The potential exists, in converting to COM, for ash deposition and ash slagging (if the furnace tempera- ture is not maintained below the ash fusion temperature). Bridging of molten ash between the tubes leads to impaired heat transfer and possible boiler derating. Soot blowers will generally prevent these problems; although some existing oil- fired boilers have soot blowers, additional soot blowing capacity may be needed to accommodate COM combustion. Since most COM combustion studies have found air or steam atomization preferable to mechanical atomization, the burners may have to be modified or (in some cases) replaced. Consideration in burner design or modification is also likely to facilitate switching from COM-to oil- firing with maximum flexibility. Combus- tion of COM requires a larger combustion space relative to oil to allowforthe longer residence time needed for complete combustion of the coal particles. Refitting for COM combustion will require provi- sions for a bottom ash handling facility in most cases since units designed for oil- and gas-firing do not often have bottom ash removal capability. In addition to modifications made directly to the boiler, other changes to the facility may be required in converting to COM. These changes are primarily associated with the fuel handling system (including pumps, piping, valves, and flow measurement devices). Storage tanks for COM received from a centralized off-site preparation plant may have to be modi- fied to include agitators and temperature control. And new storage equipment will be required if dual fuel capability (COM and oil) is desired. Economic Considerations Several economic factors will also affect the desirability of conversion to COM firing. Paramount among these considerations of the use of COM in general is the cost of coal relative to the cost of oil. Currently, widespread use of COM is limited by the low price of oil. And the increased use of COM may be limited by the advent of coal/water mixtures as a viable fuel option. Although the economics of conversion for a particular boiler may be somewhat complex, particularly for utility applica- tions, the following factors are important in every case: capital availability; boiler modifications required; difference be- tween oil (or gas) and coal prices and the predicted rates of escalation of the fuel costs; security of oil or gas supply; availability, composition, and price of COM; boiler size and capacity factor; remaining useful life of the boiler; and emission controls required. The economics of conversion are also affected by the composition of the COM (percent coal, sulfur content and ash content), the COM fuel cost, and the source of the COM (on- or off-site ------- centralized preparation plant). COM fuel cost is determined primarily by the cost of coal and oil, which account for a bout 85% of the fuel cost on a Btu basis. COM Emissions Generally, when an oil-fired boiler is to be converted to COM combustion, the uncontrolled emissions of particulates and NO, will be greater for COM-firing than for oil-firing. The increased emis- sions result from the contribution of coal ash and nitrogen to the combustion emissions. Emissions of SO2 from COM combustion may be less than or greater than oil-only SOz emissions, depending on the relative sulfur contents of the COM and the oil. Trace element emissions (except Ni and V) from COM combustion may also be greater than those for oil- only firing. Paniculate Emissions Measured emissions of particulates (fly ash) from boilers firing COM are a function of the ash content of the COM and, to some degree, the amount of ash deposition in the boiler. Since the ash content of residual oil is negligible when compared to the ash content of coal, the COM ash content depends primarily on the coal ash content and the percentage of coal used in preparing the COM. Although most test data suggest that most of the COM ash (80-90% plus) is emitted as fly ash, some COM combustion tests have shown significantly lower fly ash emissions. Ash deposition in the boiler may cause the reduced measured particulate emissions. SOz Emissions SO2 emissions from COM combustion are a direct function of the fuel sulfur content. The COM sulfur content is in turn determined by the sulfur contents of the coal and oil used to make the COM and the relative percentage of each in the COM. At least 95% of the fuel sulfur is typically emitted as SO2. But COMs made with coal that has highly alkaline ash may emit slightly less than 95% of the fuel sulfur, since the alkaline ash retains some of the fuel sulfur. Depending on the relative sulfur contents of the oil and coal, and the percentage of each fuel in the COM, SOz emissions from the COM may be greater or less than SOz emissions from the respective oil. /VOx Emissions Emissions of NOx from boilers firing COM are more difficult to quantify for all potential applications than are SOz and particulate emissions. For any particular boiler, NOx emissions can vary not only with COM fuel composition, but also with the amount of combustion air (excess air) and, in some cases, with boiler load. Properties of the COM fuel that influence NOx emissions are: nitrogen content of the fuels used to make the COM, percent coal in the COM, and, possibly, the presence of water in the COM. For a given fuel composition, NOX emissions can vary significantly from boiler to boiler due to differences in burner and furnace design and the use of combustion air preheat. NOx emission data from several differ- ent COM combustion tests indicate several general trends: • An increase in NOx emissions at increased excess air levels. • An increase in NOx emissions with increased COM nitrogen content. • Higher NOx emissions from COM combustion than from combustion of corresponding oil. • NOx emissions of 172-301 ng/J (0.4- 0.7 lb/106 Btu) in most tests, al- though some tests showed emissions of 430 ng/J (1.0 lb/108 Btu) or greater. Trace Element Emissions Trace elements i n the COM fuel exit the boiler either with the bottom ash or with the flue gas, if there is not significant ash deposition in the boiler. Most of the trace elements emitted with the flue gas are associated with the fly ash, though some may remain in the vapor phase. The amount of trace elements emitted from a particular boiler depends on: combustion temperature, fuel feed mechanism, characteristics of the flue gas, and COM properties (trace element concentration). The combustion temperature deter- mines the extent to which specific trace elements are volatilized and thus the extent to which they may be emitted with the fly ash or flue gas. The fuel feed mechanism influences the partitioning of non-combustible trace elements be- tween the bottom ash and the fly ash. The temperature of the flue gas affects the relative amounts of volatile trace elements which are emitted condensed on the fly ash particles compared to being emitted as a vapor. The concentration of trace elements in the coal and oil and the relative amounts of coal and oil used to prepare the COM determine the concentration of trace elements in the COM. Data on three coals show that coal has higher concentrations of As, Be. Cr, Hg, and Se. Residual oil has higher concentrations of Cd. Ni, and V. Pofynuclear Organic Material (POM) The amount of POM emitted from any combustion source depends on the formation and the transformation mech- anisms of the POM. POM is formed in the combustion zone either by the breakdown of larger molecules or by the building up of smaller ones. Evidence indicates that POM forms in the vapor phase and later condenses on flue gas particulates. POM formation is related to combustion efficiency, and POM transformations are related to boiler and downstream flue gas temperatures. When properly fired, oil- only combustion has been shown to contribute almost no POM emissions to the environment, while coal-only com- bustion produces POM emissions in unpredictable patterns. Therefore, the POM emissions from COMs cannot be related to the fuel content of any particular component. Applicable Control Technologies Particulate Control Technologies Particulate control technologies used on coal-fired boilers are: electrostatic precipitators (ESPs), fabric filters, venturi (wet) scrubbers, side stream separators, and mechanical collectors. Side stream separators and mechanical collectors are used only on industrial coal-fired boilers. Most industrial and utility oil-fired boilers are equipped with ESPs or have no particulate matter control device. ESPs are primarily used on utility boilers, while most oil-fired industrial boilers are uncontrolled. Mechanical collectors and venturi scrubbers are not often used to control particulate emissions from oil- firing due to the relatively small size of the oil fly ash (generally less than 2 /jm) and the inefficiency of these collectors in removing small particles. High particulate control efficiencies (98% or greater) have been widely demonstrated with ESPs, fabric filters, and wet scrubbers. In general, these technologies can reduce fly ash emissions to 43 ng/J (0.1 lb/106 Btu) or less, comparable to controlled and, in many cases, uncontrolled fly ash emissions from oil-fired boilers. Fabric filters are generally the most effective of the three technologies. Bag blinding (caking of moist particulates on the filter bag), mentioned in some literature as a ------- potential problem due to the hygroscopic nature of the fly ash, has not been observed in limited COM experience. In fact, the fly ash from COM combustion appears to have characteristics similar to coal fly ash. Coal fly ash has been shown to be highly suitable for fabric filter collection. With proper design (i.e., adequate specific collection plate area), ESPs can be as efficient as fabric filters. There is a lack of resistivity data on COM fly ash, however; these data may be particularly important for boilers firing a COM made from low sulfur coal. It is likely that the resistivity properties of the coal fly ash will dominate the characteristics of the COM ash. Wet scrubbers are somewhat less effective than fabric filters or ESPs, especially for collecting submicron particles. Venturi scrubbers are generally the most efficient type of scrubber for particulate removal. Wet scrubbers, however, have two main disadvantages: they result in wet solids sludge and may have liquid waste impacts; and the flue gas becomes saturated, reducing plume buoyancy and adversely affecting disper- sion of gaseous pollutants(SO2andNO,). Combined SOz /Particulate Control Technologies The combined SO2/particulate control technologies in use are: wet flue gas desulfurization (FGD), spray drying FGD, and fuel cleaning and blending prior to combustion. Technologies still being developed include: dry injection of sodium-based compounds, and alkaline fuel additives. Except for alkaline fuel additives, these technologies can be used to simultaneously control SOz and particulates. In general, the technology exists to control SO2 from COM combustion to levels equal to or lower than those from existing oil-fired boilers. FGD systems, capable of achieving at least 90% SOz removal, are the most effective SOz control technologies. Significant SOz control can also be obtained by: dry FGD (spray drying, dry injection) to achieve 50% or more SOz removal from a low- sulfur COM; or COM prepared with physically cleaned coals, hydrodesulfur- ized oil, or naturally occurring low-sulfur fuels. Blending cleaned fuels (or naturally occurring low-sulfur fuels) to make the COM avoids boiler-site solid or liquid waste impacts associated with FGD control. Wet scrubbers are su itable for combined SOa/particulate removal. Combined SOz/particulate control with this system favors venturi scrubbers, possibly pre- ceded by a mechanical collector. Sub- stantial combined SOz/particulate control is also achievable with spray drying systems (including a fabric filter or an ESP) and with dry injection of sodium- based alkali compounds into a fabric filter. The disposal of highly soluble sodium-based wastes from dry injection may present more serious solid waste dis- posal problems than the disposal of cal- cium-based wastes from spray drying or wet FGD systems. Physical coal cleaning (PCC) can reduce the ash content of the coal as well as reducing its sulfur content. But, depending on the percentage of coal in the COM and the final ash content of the cleaned coal, combustion of COM made from PCC will still likely require additional control to reduce particulate emissions to levels typical of oil combustion. /VOx Control Technologies Control technologies for NOX can be divided into: combustion modification controls, and post-combustion controls. Combustion modification includes: low excess air (LEA) operation, staged combustion, flue gas recirculation (FGR), and low-NOx burners. Post-combustion controls include ammonia injection ("thermal De-NOx") and flue gas treat- ment. Despite the use of what are generally accepted as the most efficient techno- logies applicable to COM units, NOX emissions from COM combustion may still be greater than those from combustion of typical No. 6 residual oils. Low-NO* burners are the most effective of the candidate NO, control technologies examined in this study. Limited data indicate that low-NOx burners can reduce NO, emissions to oil-only levels or less. However, for high nitrogen content COMs, NOX emissions may still be above the levels prescribed in some emission regulations, particularly in states like California which have stringent NO, regulations. In addition, the development status of low-NOx burners for commercial utility and industrial boilers firing COM is questionable. Staged combustion is a relatively effective NOx control technology. This technique reduces fuel NOxformation and is thus particularly suitable for high nitrogen content COMs, which produce substantial amounts of fuel NOx when combusted. Staged combustion has reduced NO, emissions 30-40% during COM combustion tests. Operation at low excess air (LEA) levels is primarily effective in reducing thermal NO,. Available data from LEA tests during COM combustion show NOx reductions of 10-20%. Thus, more than just LEA control should be required for high nitrogen COMs to reduce NOx emissions to the levels typical of oil combustion. Note that the effects of variables (e.g., COM properties, the presence of water in the COM, combustion air preheat temperatures, and boiler heat release rate) on NOx emissions from COM combustion are not fully quantifiable based on existing data. Also, boiler-to- boiler variations in burner and furnace design may significantly vary NO, emis- sions from combustion of a given COM. Trace Element Control Technologies Technologies with the greatest degree of fine particulate control are the most efficient for trace element collection, since many of the trace elements tend to be enriched on the smaller fly ash particles. Thus, fabric filters achieve the greatest degree of trace element control, followed by ESPs and wet scrubbers. In addition, physical coal cleaning and oil hydrodesulfurization can reduce trace element concentration in the fuel prior to combustion. Cost Impact of Control Technologies The cost impacts of various particulate, SO2, and NOx control technologies for boilers firing COM were evaluated in terms of: capital costs, annual operating and maintenance costs, annualized costs, and cost effectiveness. (For this study, cost effectiveness is defined as the annualized cost of control divided by the tons of pollutant removed.) The impacts of various particulate and SOz control technologies were evaluated for three COM boiler sizes and four COM fuel types. The boiler sizes examined were: 8.8, 73, and 205 MW (30, 250, and 700 x 106 Btu/hr). The COM fuel types for which the control technologies were costed were selected from typical coal and oil compositions. Uncontrolled SOz and particulate emission rates were calculated assuming that: all of the fuel sulfur is emitted as SOz, and 90% of the COM ash is em itted as fly ash. Each COM fuel was assumed to be a 50:50 mixture of coal and oil. The costs of particulate and SC>2 control technologies given in the report are based primarily on the techno- logy costs used in EPA's development of Industrial Boiler New Source Performance Standards (NSPS). ------- Comparison of FGD System Costs The costs of various FGD systems are compared for units firing low- and high- sulfur COM. The costs are based on 70% S02 removal [a controlled SOz emission level of 168 ng S02/J (0.4 lb/106 Btu)] and 60% annual capacity factor. The relative costs of the three FGD systems for specific applications may be altered due to site-to-site variations in SOa removal, boiler capacity, reagent costs, or availability of existing equipment to reduce retrofit costs. For boilers firing low-sulfur COM, dry injection FGD has the lowest capital costs up to a boiler size of about 200 x 106 Btu/hr. However, wet FGD has the lowest capital cost of the three systems for larger boilers. The result is due primarily to the increased paniculate matter collection associated with the use of a fabric filter in the spray dryer and dry injection systems. The lower paniculate emission control levels achievable with fabric filters, relative to a wet FGD system used for combined SOz/particulate removal, result in higher waste disposal costs. An ESP would likely be required upstream of the wet FGD to achieve the same particulate emission level. Similar findings were made for high-sulfur COM. In comparing the annualized costs for three FGD systems, all processes were relatively close for low-sulfur COM. Wet FGD had the lowest annualized cost for boilers above about 200 x 106 Btu/hr. Wet FGD was also the least expensive system for high sulfur COM, while dry injection had the highest annualized costs. Cost Comparison of Low Sulfur COM With Wet FGD The annualized costs of using three low-sulfur COMs were calculated relative to the cost of using a 2.3% sulfur COM. These costs are compared (in Figure 1) to the costs of wet FGD at 70% SOz removal. Emissions of SOz for each are less than 430 ng/J (1.0 lb/106 Btu}. Wet FGD is more expensive than low- sulfur COMs with 50% coal. However, the 40% coal COM (made with 0.7% S coal and 0.5% S oil) is more expensive than FGD for boilers larger than about 300 x 106 Btu/hr. Of course, the boiler size where FGD is less costly than low sulfur COM depends on boiler capacity factor. Although both FGD and annual COM costs depend directly on boiler capacity factor, COM costs are generally affected more strongly by changes in capacity factor. FGD should be less costly than low-sulfur COM only for large boilers operated at relatively high annual capacity factors (greater than about 0.6). Low-sulfur COM and an ESP or a fabric filter will also provide simultaneous SOz/particulate control. The combined annual cost of low-sulfur COM and a fabric filter is compared to the costs of wet and dry FGD in Figure 2. Emission control levels achievable with these systems are: SOz—nominal 430 ng/J (1.0 Ib SOz/106 Btu); and particulate—21.5 ng/J (0.05 lb/106 Btu) (dry FGD + fabric filter), and 43 ng/J (0.1 lb/106 Btu) (wet FGD). Figure 2 shows the 50% coal low- sulfur COM/fabric filter combination to be the least expensive combined S02/ particulate control technique up to a boiler size of about 325 x 106 Btu/hr. For larger boilers, wet FGD is the least expensive alternative. Both wet FGD and spray drying are less costly than the 40% coal low-sulfur COM/fabric filter com- bination. 3,500 3,000 2,500 CD ^ 2,000 »•» «i » o O .§ 7,500 /,ooa 500 Wet FGD; 70% SO2 removal: 2.3% S COM IB-1) 0.92 Ib SOi/10* Btu 40% COM (0.7% S coal/0.5% S oil); 0.77 Ib SO2/10*Btu 50% COM 10.7% S coal/0.5% S oil); 0.83 Ib SOz/10*Btu 50% COM (0.7% S coal/0.87% S oil); .0.99 Ib SO*/10* Btu Note: Low sulfur COM costs calculated relative to cost of 2.3% S COM (B-1) 'COM Wet FGD COM COM 60% Annual Boiler Capacity Factor • Double Alkali Wet FGD System (Combined 502 /Particulate Control) 0 (01 29 (100) 59 (200) 87 (300) 116 (400) 145 (500) 174 (600) 203 (700) 232 (800) BoilerSize, MW(10*Btu/hr) Figure 1. Comparison of wet FGD and low-sulfur COM annualized costs. 5 ------- 8 o I " 5500 r 5000 4500 4000 3500 3000 2500 2000 1500 1000 500 Low-Sulfur COM/Fabric Filter FGD Based on 60% Boiler Capacity Factor 40% Coal Low Sulfur COM / fO. 7% S Coal/0.5% S Oil) ' Wet FGD 70% SO2 Removal - 50% Coal Low Sulfur COM (0.7% S Coal/0.5% S Oil) 0 29 59 87 116 145 174 203 232 fO) (100) (200) (300) (400) (500) (600) (700) (800) Boiler Size, MW (10* Btu/hr) Figure 2. Comparison of annual/zed costs of alternatives for combined SOz /paniculate control. ------- M. E. Kelly, R. M. Parks, andJ. H. E. Stellingare with Radian Corporation, Durham, NC 27705. Robert E. Hall is the EPA Project Officer (see below). The complete report, entitled "Control of Criteria and Non-Criteria Pollutants from Coal/Oil Mixture Combustion," (Order No. PB 83-247 247; Cost: $19.00. 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: Industrial Environmental Research Laboratory U.S. Environmental Protection Agency Research Triangle Park, NC27711 United States Environmental Protection Agency Center for Environmental Research Information Cincinnati OH 45268 Postage and Fees Paid Environmental Protection Agency EPA 335 Official Business Penalty for Private Use $300 PS 0000329 ------- |