|
|
||||||||
a School of Natural Resources, The Ohio State University, Wooster, OH 44691
b School of Natural Resources, The Ohio State University, Columbus, OH 43210
c Department of Crop and Soil Sciences, Pennsylvania State University, University Park, PA 16802
d Sierra Nevada Research Institute, University of California, Merced, CA 95344
e Department of Civil and Environmental Engineering and Geodetic Science, The Ohio State University, Columbus, OH 43210
* Corresponding author (dick.5{at}osu.edu)
Received for publication May 24, 2004.
| ABSTRACT |
|---|
|
|
|---|
and Ca. Concentrations of Ag, As, Ba, Cd, Cr, Hg, Pb, and Se (except for ASTM Se in two samples) were below drinking water standards in both ASTM and TCLP leachates. Total toxicity equivalents (TEQ) of dioxins, for two FGD products used for mine reclamation, were 0.48 and 0.53 ng kg1. This was similar to the background level of the mine spoil (0.57 ng kg1). The FGD materials were mostly uniform in particle size. Specific surface area (m2 g1) was related to particle size and varied from 1.3 for bed ash to 9.5 for spray dryer material. Many of the chemical and physical properties of these FGD samples were associated with the quality of the coal rather than the combustion and SO2 scrubbing processes used.
Abbreviations: AFBC, atmospheric fluidized bed combustion ASTM, American Society for Testing and Materials CCE, calcium carbonate equivalent, FGD, flue gas desulfurization LIMB, lime injection multistage burner PFBC, pressurized fluidized bed combustion, TCLP, toxicity characteristic leaching procedure TEQ, total toxicity equivalents
| INTRODUCTION |
|---|
|
|
|---|
The burning of coal converts sulfur-bearing impurities to gaseous SO2, an atmospheric pollutant and a precursor to acid rain. Accordingly, the Clean Air Act Amendments of 1990 limit the total amount of SO2 that can be released to the atmosphere. Much of the coal in the eastern United States contains sulfur at levels high enough that its combustion will cause utilities and other users to be out of compliance without remedial action. This problem has spurred the development of various types of scrubbing processes to convert SO2 from flue gases into solid products for disposal or beneficial reuse. These FGD techniques are characterized as being either wet or dry, and they may differ substantially in the sorbents used and the products created (Electric Power Research Institute, 1999).
Because of our heavy dependence on coal as an energy source, a large amount of coal combustion products is created each year. According to the 2001 survey by the American Coal Ash Association (American Coal Ash Association, 2004a), more than 118 million Mg of coal combustion productsincluding fly ash, bottom ash, boiler slag, and FGD materialswere produced. The quantity of coal combustion products created each year is slightly less than sand, gravel, and crushed stone and more than Portland cement and iron ore. In 2001, the amount of FGD material produced was 28.5 million Mg or 24.2% of the total coal combustion products generated. Due to an expected increase in the application of scrubbing technologies, the amount of FGD material that will be generated should grow during the next few decades. Approximately 7.58 million Mg (26.6% of the total FGD produced in 2001) were used in various ways. Commercial products containing FGD materials include wallboard, flowable fill material, cement additives, and aerated concrete block. Flue gas desulfurization materials have also been successfully used for waste stabilization, roadway and runway construction, mine reclamation, and as agricultural amendments (Electric Power Research Institute, 1999; Dick et al., 2000).
Substantial amounts of information on the properties of coal fly ash and/or bottom ash are available in the literature. Of note are papers published by Adriano et al. (1980), Carlson and Adriano (1993), Querol et al. (1995), Hower et al. (1996), and Martinez-Tarazona and Spears (1996). The American Coal Ash Association (Aurora, CO, USA) hosts a library and website where much additional information can be obtained (American Coal Ash Association, 2004b). In contrast, much less information exists on the properties of FGD products. Punshon et al. (1999) and Dick et al. (2000) give basic descriptions of the production, uses, and characteristics of FGD products. Some specific information on the chemical and physical characteristics of FGD products can also be found in Hower et al. (1997)(1999), Gomes et al. (1998), Armesto and Merino (1999), and Steenari et al. (1999). These studies, however, are incomplete and often focus on only one type of FGD product.
The continued development of beneficial uses of FGD products requires that we have detailed knowledge of their chemical and physical properties. This information is important both from the standpoint of creating effective new products as well as preventing the spread of unsafe materials in the environment. For agricultural uses, especially, it is important that the potential risks of applying FGD as a soil amendment (e.g., as a fertilizer, liming agent, soil physical conditioner, etc.) be evaluated.
Dry FGD systems often remove SO2 and coal ash simultaneously, and the final material is a mixture of unreacted sorbent, fly ash, and sulfur reaction products. Flue gas desulfurization materials are also affected by the various technologies used to remove the SO2 from the flue gases. This paper describes the chemical and physical characteristics of 59 dry FGD samples collected from 13 locations representing several different SO2 scrubbing technologies.
| MATERIALS AND METHODS |
|---|
|
|
|---|
|
Lime Injection Multistage Burner (LIMB) Samples
A calcium-based sorbent is injected into the boiler, where it calcines to CaO and/or MgO, and reacts with SO2 and O2 in the combustion gases to produce CaSO4. Temperatures at the injection point approach 1260°C and reaction occurs between this temperature and about 870°C. The reaction product and unspent sorbent are collected with the fly ash in electrostatic precipitators or other particulate emission control systems.
Fluidized Bed Combustion (FBC) Samples
A calcium-based sorbent (usually limestone or dolomite) and crushed coal are introduced together into the boiler bed where they are fluidized or suspended by jets of air. Reaction of sorbent with SO2 occurs in the boiler at temperatures usually less than 870°C. Two FGD product streams are produced: a heavier, granular bed ash material and smaller particles suspended in the flue gas. The smaller particles are removed with the particulate emission control system. Both products contain reaction products and unspent sorbent as well as either conventional bed or fly ash. Fluidized bed combustion systems may operate at either atmospheric pressure or at higher pressures.
Spray Dryer Samples
The sorbent, usually a slurry of hydrated lime, is injected into the flue gases in a separate scrubber vessel located downstream of the boiler. The scrubber vessel increases the residence time of flue gas and thus increases the time period for sorbent to react with SO2. Reaction products, unspent sorbent, and fly ash are removed together in the particulate emission control system.
Other Samples
Several samples were obtained from places within the coal burning and scrubbing facilities that were not representative of the final product normally created or they were from a scrubbing process that was represented by only a single sample. These samples are included in this report because they represent another range of samples that may be encountered when flue gases are scrubbed to remove SO2.
Total Chemical and Dioxin Analysis
When conducting total elemental analyses of the dry FGD product samples (and the insoluble residue samples mentioned later in the Materials and Methods section), we also included standard fly ash samples and compared our values for these standard samples with certified values reported by the National Institute of Standards and Technology (NIST). If values were outside of accepted levels (usually set at 10% relative percent deviation), the analyses were repeated until acceptable values were achieved.
Total chemical analysis of the solids was determined using duplicate, 100-mg FGD samples digested with an aqua regiaHF mixture in Teflon decomposition vessels. The vessels were placed in stainless steel digestion bombs and heated at 110°C for 40 min. Digested samples were then mixed with excess H3BO4 and diluted to 100 mL total volume with distilled H2O. Digests were analyzed for Al, Ba, Be, Ca, Fe, K, Li, Mg, Mn, Mo, Na, P, S, Si, and Sr using inductively coupled plasma emission spectrometry with a PS2000 instrument (Leeman, Los Angeles, CA). Concentrations of As, Cd, Co, Cr, Cu, Ni, Pb, and Se in the digests were measured with a 4100-ZL graphite furnace, atomic absorption spectrophotometer equipped with a Zeeman background corrector and L'vov-platform atomization (PerkinElmer, Wellesley, MA). Electrode-discharge lamps were used for As and Se analysis, and hollow cathode lamps were used for all other elements. Zinc concentrations were measured by atomic absorption using a Techtron atomic absorption spectrophotometer AA6 instrument (Varian, Palo Alto, CA).
Total B concentrations were determined on 100-mg samples by a custom method (USGS sodium peroxide sinter procedure). Samples were placed in Pt crucibles, mixed with 0.4 g of Na2O2, heated at 445°C for 30 min and allowed to sinter, and then removed from the furnace and cooled rapidly. Fourteen milliliters of distilled H2O were added to the sinter cake and the mixture was stirred for 10 min. Six milliliters of 6 M HCl were then added to the solution and stirred until the sinter cake was completely dissolved. Solutions were diluted to 100 mL total volume with distilled H2O and heated in a drying oven for 1 h at 100°C. Boron concentrations were measured by inductively coupled plasma emission spectrophotometry using the Leeman PS2000 instrument.
Dioxin (defined here as the totality of 7 dioxins and 10 furans from a much larger family of similar but less toxic compounds of environmental concern) analyses were also conducted on selected FGD samples obtained from a pressurized fluidized bed combustion unit (American Electric Power Plant, Brilliant, OH) and an atmospheric fluidized bed combustion facility (General Motors, Pontiac, MI). To gain approval of these FGD materials, used either alone or mixed with yard waste compost to help remediate a toxic mine spoil, we were required to conduct a dioxin analysis of the materials. The yard waste compost was obtained from a commercial composting facility (Earth N Wood) located in North Canton, OH. The feedstock for the compost was comprised of tree leaves, grass clippings, mulch, and other yard wastes. Spoil samples from the mine site were obtained after grading had been completed but before final reclamation by addition of FGD amendment. Samples of the FGD, compost, FGD plus compost mix (approximately 5:2 by weight or 1:1 by volume of FGD to compost), and mine spoil were placed into bags supplied by Triangle Laboratories (Research Triangle Park, NC), and sent to the laboratory by overnight mail. The samples were then subjected to full screen analyses for the presence of polychlorinated dibenzo-p-furans and dibenzofurans by high-resolution chromatographyhigh-resolution mass spectrometry using Method 8290 (USEPA, 1994). Total dioxin toxicity equivalents (TEQ) (in ng kg1) were calculated based on the presence of various dioxin congeners according to a worksheet procedure prescribed by the Ohio Environmental Protection Agency (Columbus, OH).
Insoluble Residue Content
Insoluble residue contents were measured by mixing weighed samples (1020 g) of FGD product with sodium acetateacetic acid buffer (pH 5.0) to form thick slurries. The slurries were treated with 30% H2O2 to convert CaSO3·0.5H2O (if present) to the more soluble CaSO4·2H2O. Following oxidation and heating at 60°C to decompose excess H2O2, the samples were quantitatively transferred to cellulose dialysis bags and submerged in sodium acetateacetic acid buffer (pH 5.0) for 4 to 6 d to remove acid-soluble materials. The samples were then dialyzed against distilled H2O to remove excess salt and to dissolve residual CaSO4·2H2O. Removal was considered complete when a negative test for dissolved SO42 was obtained with BaCl2. The samples were quantitatively transferred to tared beakers, dried at 100°C, and weighed to determine residual ash content. Subsamples (100 mg or less if sample size was limiting) of the residual ash were completely digested with an aqua regiaHF mixture in Teflon decomposition vessels and analyzed for total elemental concentrations as described previously. Total elemental concentrations in the insoluble residues were subtracted from the total original sample concentrations to determine the relative proportion of an element due to coal ash residual versus that introduced with the sorbent or by subsequent reaction of the sorbent and flue gases.
Leachate Analysis
Standard 20:1 (w/w extractant to sample), 18-h leachings were performed using the deionized water leaching procedure (Method D 3987-85; American Society for Testing and Materials, 1990a) and the toxicity characteristic leaching procedure (TCLP) (USEPA Method 1311). The TCLP method uses dilute acetic acid (5.7 mL glacial acetic acid diluted to 1 L with deionized water (final pH = 2.88 ± 0.05) as the extractant (Code of Federal Regulations, 1991, p. 6681). For both procedures, 2 L of extractant and 100 g of FGD product were mixed in a Teflon bottle. Bottles were shaken for 18 h at 25°C on an end-over-end rotary shaker rotating at 30 rpm. Leachates were filtered (0.22-µm openings) and analyzed for pH, total dissolved solids, four anions
by ion chromatography, and 30 additional elements by ICP and graphite furnace atomic absorption spectrophotometry. A total of 24 samples were analyzed by the ASTM method and six samples by the TCLP method. All analyses were replicated twice.
Paste pH, Total Neutralizing Power, and Available Lime
Duplicate paste pH values were determined by mixing equal volumes of FGD product and distilled water in a 70-mL cup. The mixture was stirred, allowed to equilibrate for 30 min, and then stirred and equilibrated again. Paste pH was measured by using a glass electrode with an EA920 expandable ion analyzer (Orion Research, Beverly, MA).
Total neutralizing power was determined in triplicate as calcium carbonate equivalents according to ASTM C 602-90 (American Society for Testing and Materials, 1990b). Samples were heated with excess standard HCl and back-titrated with standard NaOH. Reagent-grade CaCO3 was used as a standard reference material.
Available lime index was measured for selected samples using ASTM C 25-90, Section 33 (rapid sugar method) (American Society for Testing and Materials, 1990c). Samples were slaked with water and the lime was solubilized by reaction with sugar to form calcium sucrate. The calcium sucrate was then determined by titration with standard HCl.
Physical Methods
Particle size distributions were measured using the hydrometer and dry sieving methods of ASTM D 422-63 and Dispersion Apparatus A (American Society for Testing and Materials, 1990d). Independent particle density values were employed for all size distribution calculations. Results were expressed as percent of particles finer than 0.025 mm (or for coarse samples the percent of particles finer than 0.25 mm). The coefficient of uniformity (Cu) was calculated as Cu = D60/D10, where Dx is the effective particle diameter separating the finer "x" percent of the particles from the coarser (100 x) percent (Skopp, 2000).
Particle densities were measured in duplicate using a Model 1305 multivolume gas pycnometer (Micromeritics, Norcross, GA) on approximately 4 cm3 of weighed sample. The instrument was initially calibrated by placing steel balls in the sample chamber. The reported particle densities represent the mean of three instrument measurements for each replicate. A standard reference sample (NIST 1825) was measured at the beginning of each operating period and after every 20th sample.
Specific surface areas were measured using nitrogen absorption by the continuous flow method (ASTM D 4567) with a Micromeritics Flowsorb II 300 instrument (American Society for Testing and Materials, 1990e). The instrument was calibrated at the beginning of each operating period by injecting a known volume of analytical grade nitrogen gas per ASTM D 4567. Two standard reference materials (NIST 8570 and 8571) were analyzed at the beginning of each operating period and after every 20th sample. Quantities of both standards and samples were adjusted to give surface areas in the range of 0.5 to 25 m2 as suggested by the instrument manufacturer. Sample materials were analyzed in triplicate or until individual analyses were within ±10% of the mean values following removal of any outlying data points.
| RESULTS AND DISCUSSION |
|---|
|
|
|---|
|
|
Although formulated to regulate land application of biosolids, the 40 CFR 503 regulations (USEPA, 1989) have become the de facto standard for land application of many materials that contain potentially hazardous metals (Dick et al., 2000). The regulations can serve in a comparative way for assessing the suitability of FGD products for land application. All FGD products (Table 3) met the Part 503 criteria (mg kg1) for Cd (85), Cr (3000), Cu (4300), Pb (840), Ni (420), Se (100), and Zn (7500). For As, 23 samples exceeded the 75 mg kg1 limit. In general, the greatest As concentrations were present in the FGD products from the LIMB process (Table 3) although this result was probably due more to the source of coal burned than the combustionscrubbing combination used. Solid-phase concentrations of Hg in the FGD products were measured in only a limited number of samples, but in these samples, the concentrations were below our analytical detections limits of 5 mg kg1 and this level was approximately an order of magnitude below the Part 503 criterion for Hg (57 mg kg1). Low concentrations (<0.0002 mg L1) of dissolved Hg in leachates from the TCLP procedure (next section) also indicate that FGD products are not a threat for Hg contamination. For Mo, one sample (Sample 16) had a concentration of 63.7 mg kg1 and two other LIMB samples, one duct injection sample and one spray dryer sample also had concentrations that exceeded 40 mg kg1. The remaining samples all had concentrations that were below 29.6 mg kg1 with the majority (33 samples) having concentrations of 10 mg kg1 or less.
The ash residues, following removal of acid- and water-soluble components, were dominated by Al, Fe, and Si. This is consistent with the mineralogy of coal ash (Hower et al., 1996). Plots of total element concentration versus residual ash content in the FGD samples yielded patterns (Fig. 1) that usually suggested a partitioning of most elements into either the residual ash or soluble components. For example, Al consistently increased with residual ash content. For such elements (i.e., Al, Ba, Cr, Fe, K, Si, and V) all points were closely packed about a sloped line that was determined by the concentration of the element in the residual ash sample. Thus the concentrations of these elements in the residual ash were affected by coal source but not the scrubbing technology used. For Be, Li, and Pb, the slopes of element concentrations versus residual ash content were different for different scrubbing technologies, clearly suggested a dependency on operating conditions within a given facility and coal composition. For B (Fig. 1) and some other elements (i.e., As, Co, Cu, Mo, Ni, P, Se, and Sr), concentrations generally increased as residual ash contents increased except for the spray dryer samples. Concentrations of these elements in the spray dryer samples either were unrelated to residual ash content or decreased slightly indicating the source was both from the coal burned and the sorbent introduced for scrubbing. Cadmium levels in the LIMB and spray dryer samples showed increased and decreased concentrations, respectively, as residual ash contents increased. Manganese concentrations increased with residual ash content for all samples except those from the LIMB process. The reasons for this distribution are not known.
|
ASTM and TCLP Leachates
Element concentrations in leachates obtained by the ASTM and TCLP leaching procedures are summarized in Tables 4 and 5. Results are presented as mean (±SD), range, and coded high value for all samples and thus are not separated by FGD scrubbing technology. For those parameters where many samples were below the detection limits, a mean was not calculated but the number of samples below a given limit is noted.
|
|
The leachates also contained high concentrations of total dissolved solids composed mainly of Ca, SO24, and SO23 derived by dissolution of the reaction products CaSO4 and CaSO3. The leachates generally exceeded secondary drinking water standards for both total dissolved solids (500 mg L1) and sulfate (250 mg L1). Element concentrations were usually greater with the TCLP acid extraction than with the ASTM water extraction. With the exception of Sample 22 (EDG-LIM-14), sulfite
was found only in leachates from duct injection and spray dryer samples (those FGD processes that occur under more oxygen limiting conditions). Some leachates from the PFBC samples (i.e., Samples 5153) were also high in Mg due to the use of dolomite as a sorbent in the scrubbing process.
Concentrations of eight trace metals (Ag, As, Ba, Cd, Cr, Hg, Pb, and Se except for ASTM Se in Samples 4 and 53) were below drinking water standards for both the ASTM and TCLP leachates. Mercury was below the detection limit (<0.0002 mg L1) in all of the ASTM and TCLP leachates. Concentrations of trace metals in the ASTM and TCLP leachates, other than those already mentioned in this paragraph, were also generally low for all samples. Most of these metals have low solubility because (i) they are associated with the relatively insoluble fly ash portion of the FGD product rather than the more soluble fraction composed of unspent sorbent and SO2 reaction products or (ii) the high pH of the material limited solubility.
Dioxin Toxicity Equivalents
Dioxins are released into the environment from many sources such as automobile exhaust, industrial waste incineration, and as unwanted byproducts in various chlorinated chemical formulations. It has been suggested that the combustion of coal also generates dioxins that are incorporated into FGD products. For this study, samples were collected for dioxin analyses from two high volume FGD products that were used, either alone or mixed with yard waste compost, for reclamation of an abandoned surface coal mine site. A spoil sample from the mine site was also collected to serve as an indication of the background level before adding any amendment to the site.
Total toxicity equivalent (TEQ) values of dioxin (Table 6) were very low (0.48 and 0.53 ng kg1) for the two FGD samples analyzed, essentially being equal to the background value in the mine spoil (0.57 ng kg1). These concentrations are also much below the 4 ng kg1 urban soil background level reported by Lorber et al. (1998). The 2,3,7,8-TCDD congener, the most studied member of the dioxin family, is a known human carcinogen and accounts for about 10% of our background dioxin risk (USEPA, 2003). This congener was not found in the spoil or the two FGD samples. Clearly, the coal, coal combustion, and subsequent scrubbing of the flue gases to remove SO2 did not lead to the creation or capture of dioxin in the FGD products examined.
|
Paste pH, Total Neutralizing Power, and Available Lime
The FGD products were highly alkaline with mean paste pH values exceeding 11.8 (if we exclude the "other" samples) (Table 7). Mean pH values for the various FGD scrubbing technologies and the range of observed values within each technology varied only slightly. The only exceptions were Samples 49, 51, and 54 in the cyclone ash group, which had slightly lower pH values (9.910.5). This result suggests that the pH of dry FGD products depends primarily on the sorbent used and secondarily on the FGD technology. The high pH values of most FGD samples are due to the presence of oxides and hydroxides of Ca and Mg. When these products are exposed to water and CO2, they will convert to carbonates by carbonation reactions and the pH will decrease.
|
The dry FGD products contain sufficient acid neutralizing potential to warrant their use as alkaline amendments in acid soils and spoils. However, because the calcium carbonate equivalency values are less than 100%, the FGD application rates needed to obtain the same pH adjustment benefit would have to be greater than that for limestone. Thus, before using any of these materials as liming agents to adjust soil (spoil) pH, they must first be tested to determine their specific CCE values and then the rates required to achieve a result similar to an equivalent amount of calcium carbonate must be calculated. The generally fine particle size as well as the mineral composition [CaO, Ca(OH)2 or MgO versus CaCO3 or CaMg(CO3)2] of most products also suggest that FGD materials will react rapidly to neutralize acidity.
Physical Properties
The physical properties reported here were measured primarily because they affect engineering uses of the FGD products. Uniformity of material and ability to pack tightly are especially important. The FGD products from the duct injection, spray dryer, and LIMB processes had greater than 800 g kg1 (80% by weight) of particles finer than 0.025 mm effective diameter, indicating very fine particles (Table 8). The fly ash particles from the fluidized bed process were somewhat coarser. Bed ash materials were very coarse and averaged 14 g kg1 by weight finer than 0.25 mm effective diameter. Uniformity coefficients (Cu) for all materials, except fluidized bed fly ash, ranged from 1.8 to 2.4. Coefficients less than approximately 2 indicate very uniform particle size distributions. Thus, FGD materials from all processes, except the bed ash samples from the fluidized bed combustion process, were uniform in particle size.
|
| CONCLUSIONS |
|---|
|
|
|---|
| ACKNOWLEDGMENTS |
|---|
| REFERENCES |
|---|
|
|
|---|
Related articles in JEQ:
This article has been cited by other articles:
![]() |
L. Chen, D. Kost, and W. A. Dick Flue Gas Desulfurization Products as Sulfur Sources for Corn Soil Sci. Soc. Am. J., August 20, 2008; 72(5): 1464 - 1470. [Abstract] [Full Text] [PDF] |
||||
![]() |
C.-M. Cheng, H. W. Walker, and J. M. Bigham Mechanisms Controlling the Leaching Kinetics of Fixated Flue Gas Desulfurization (FGD) Material under Neutral and Acidic Conditions J. Environ. Qual., May 7, 2007; 36(3): 874 - 886. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. B. Lee, J. M. Bigham, W. A. Dick, F. S. Jones, and C. Ramsier Influence of Soil pH and Application Rate on the Oxidation of Calcium Sulfite Derived from Flue Gas Desulfurization J. Environ. Qual., January 9, 2007; 36(1): 298 - 304. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Soil Science Society of America Journal | Journal of Plant Registrations | The Plant Genome | |||