Petrographic and Geochemical Anatomy of Lithotypes from the Blue

May 1, 1994 - ... of Trace Elements in Fractions after Micronization and Density-Gradient Centrifugation of High-Ge Coals from the Wulantuga and Linca...
0 downloads 0 Views 2MB Size
Download PDF
Energy & Fuels 1994,8, 719-728

719

Petrographic and Geochemical Anatomy of Lithotypes from the Blue Gem Coal Bed, Southeastern Kentucky James C. Hower,*ptDarrell N. Taulbee,? Susan M. Rimmer,t and Liria G. Morrellt University of Kentucky Center for Applied Energy Research, Lexington, Kentucky 40511, and Department of Geological Sciences, University of Kentucky, Lexington, Kentucky 40506 Received December 16, 1993. Revised Manuscript Received February 28, 1994"

Density gradient centrifugation (DGC) was applied to three benches of the low-sulfur, low-ash Blue Gem coal bed from Knox County, Kentucky. DGC samples were analyzed by proximate, ultimate, PIXE, FTIR, and coal petrographic techniques. The middle lithotype has an ash geochemistry influenced by carbonate associations, with an apparent variation in the carbonate substitution through the density range. The basal lithotype has a higher mineral content than the latter lithotype and appears to be enriched in silicate minerals. Cr, V, Co, Mn, Ti, and Zr are also enriched in the basal lithotype. A number of elements are concentrated in the high-vitrinite (>go%) range including Ga, Ge, and As; but it is less clear whether these trends are associated with the type of vitrinite within this range or are merely a function of the relative concentration of vitrinite.

Introduction The nature of the association of major, minor, and trace elements with coal has been the subject of intensive research by coal scientists (Swaine,' and references cited therein). Density gradient centrifugation (DGC) offers a technique with which ultrafine coal particles can be partitioned into a density spectrum, portions of which represent nearly pure monomaceral concentrates. DGC has been typically conducted on demineralized coals2 assuring, particularly at lower specific gravities, that the resulting DGC fractions would have very low ash contents. In order to determine trends in elemental composition, particularly with a view towards maceral vs mineral association, it is necessary to avoid demineralization. To this end the Blue Gem coal bed (Middle Pennsylvanian Breathitt Formation) from Knox County, Kentucky, was selected for study. The Blue Gem coal bed was known from a previous study3 to have a low sulfur content, low ash yield (in places less than 1% ), and high levels of minor element concentrations with significant differences between lithotypes in the middle and at the base of the coal bed. The objective of this study was to determine the petrography and chemistry, with particular emphasis on the ash geochemistry, of DGC separates of lithotypes of the Blue Gem coal bed. Experimental Section Sample Collection and Preparation. Samples were obtained from the Blue Gem coal bed from a mine site in the Barbourville 71/2-min quadrangle, Knox County, Kentucky. The coal bed (whole coal sample: KCER-5442)a t this sample locality was 70 cm thick and was sampled in a series of five benches (from + University of Kentucky Center for Applied Energy Research. 3 Department of Geological Sciences, University of Kentucky. e Abstract published in Aduance ACS Abstracts, April 1, 1994.

(1)Swaine,D.J. Trace elements in coal; Butterworths: London, 1990; 278 p. (2)Taulbee, D. N.; Poe, S. H.; Robl, T. L.; Keogh, R. Density gradient centrifugation separation and characterization of macerals groups from a mixed maceral bituminous coal. Energy Fuels 1989,3,662-670. (3)Hower, J. C.; Rimmer, S. M.; Bland, A. E. Ash geochemistry of the Blue Gem coal bed. Int. J. Coal Geol. 1991,18,211-231.

0887-0624/94/2508-0719$04.50/0

the top of the coal bed to base: KCER 5443-5447). Two of the bench samples were selected for further investigation. The selection was based on the knowledge from the study by Hower et aL3that the middle of the coal bed is low ash with relatively high CaO, FezOa, and Sr, while the base is slightly higher ash with relatively high TiO2, Zr, V, and Ni. Sample KCER-5447 represents the 10.0-cm basal lithotype of the coal bed. KCER5445 represents an 11.6-cm bench from the middle of the coal bed. An additional sample (KCER 5448; 50 cm thick) was collected at the same mine and represents a composite of the middle three lithologically-similarbenches of the coal bed. Bench samples were crushed and split following procedures recommended by ASTM standard D2013." Samples were crushed to -20 mesh for petrographic analysis, to -60 mesh for chemical analysis, and to -325 mesh for DGC processing. Chemical Characterization. Proximate, ultimate, and sulfur analyses were determined for each bench sample according to standard ASTM techniques' (Table 1). High-temperature ashes were analyzed for major and minor elements (Table 2A) and trace elements (Table 2B), following techniques described in Hower and Blandas Low-temperature ashes were obtained for the samples according to procedures described by Gluskoter.6 PetrographicCharacterization. Subsplits of the-20-mesh coals were mixed with epoxy resin and prepared into pellets for petrographic analysis and polished according to standard proc e d u r e ~ .Maceral ~ identification and point-count analysis (Table 3) followed techniques recommended by ICCP8and Stach et al.7 Density Gradient Centrifugation Separation. Maceral separation was based on the DGC method developed by Dyrkacz and co-workersgJOwith certain modifications.2 Additional modi(4)American Society for Testing Materials (ASTM). Annual book of ASTM Standards: Coal and Coke; ASTM Philadelphia, 1989. (5)Hower, J. C.; Bland, A. E. Geochemistry of the Pond Creek coal bed, Eastern Kentucky coalfield. Znt. J. Coal. Geol. 1989,11,205-226.

(6)Gluskoter, H.J. Electronic low-temperature ashing of bituminous coal. Fuel 1966,44,285-291. (7) Stach, E.; Mackowsky, M-Th.; Teichmiiller, M.; Taylor, G. H.; Chandra, D.; Teichmuller, R. Stach's Textbook of Coal Petrology, 3rd ed.; Gebruber Borntraeger: Berlin, 1982;535 pp. (8)International Committee for Coal Petrography (ICCP). Znternational Handbook of Coal Petrograph, Suppl. 2nd. Ed.; Centre National de la Recherche Scientifique: Paris, 1971. (9)Dyrkacz, G. R.; Bloomquist, C. A. A.; Horwitz, E. P. Laboratory scale separation of coal macerals. Sep. Sci. Technol. 1981,16,15711588. (10)Dyrkacz, G. R.; Horwitz, E. P. Separation of coal macerals. Fuels 1982,61,3-12.

0 1994 American Chemical Society

Hower et al.

720 Energy & Fuels, Vol. 8, No. 3, 1994 Table 1. Proximate, Ultimate, and Sulfur Forms Analysis of Blue Gem Bench Samples 5443 5444 5445 5446 5447 5448

53.04 57.95 60.47 61.13 61.50 59.40

37.81 41.02 39.13 38.18 37.03 38.99

9.15 10.3 0.40 0.69 1.47 1.61

2.32 0.80 0.63 0.64 0.61 0.78

0.01 0.01 0.01 0.01 0.01 0.01

1.00 0.04 0.01 0.01 0.01 0.06

1.30 0.75 0.61 0.62 0.58 0.71

84.93 86.18 86.71 85.64 86.32 85.63

5.79 5.68 5.38 5.47 5.37 5.55

2.04 2.09 1.81 1.92 1.85 1.94

4.69 5.24 5.47 6.33 5.84 6.08

Table 2. Geochemistry of Blue Gem Bench Samples: (A) Major Oxides; (B) Minor Elements (Ash Basis) and Chlorine (Whole-Coal Basis) by XRF and PIXE (Basis As Designated) A. Major Oxides (Ash Basis) sample 5443 5444 5445 5446 5447 5448

Si02 45.87 34.49 26.65 29.21 47.52 41.71

A1203 21.29 23.22 23.89 28.05 24.93 23.99

Ti02 0.81 0.90 1.14 10.2 1.59 0.96

Fez03 23.41 14.55 13.05 12.42 5.68 14.88

MgO 1.32 2.05 2.78 2.00 1.49 2.07

CaO 0.93 10.79 14.11 11.97 8.79 5.18

NazO 0.51 1.84 2.34 2.66 0.49 1.46

KzO 3.52 1.74 0.68 0.61 1.51 2.84

P206

0.12 0.18 0.41 0.21 0.25 0.18

SO3 0.42 8.40 13.19 9.98 5.84 5.07

B. Minor Elements by XRF (Ash Basis) and PIXE (Ash Basis unless Noted Otherwise) XRF 5443 5444 5445 5446 5447 5448

Ba 1260 6160 5440 5260 3840 2960

PIXE 5443 5444 5445 5446 5447 5448

Cr 299 302 445 510 1720 2550 Ga 74 50 0 161 874 36

co 45 80 171

388 426 105 Ge 214 95 65 173 3390 67

cu 720 630 1270 1380 560 660 As 931 175 63 35 0 205

Mn 281 403 530 381 1060 436

Mo 0 32 33 21 0

26

Y 94 198 130 404 496 182

fications included omission of the demineralization step, crushing the feed coal to -325 mesh, and adding Brij-35 surfactant to the separation media (due to the smaller particle size) to suppress particle agglomeration. Details of the DGC method are given elsewhere;2 the following is offered only as a brief description. A Beckman Model 52-21 centrifuge fitted with a 1900-mL capacity titanium JCF-Z zonal core rotor was used for all DGC separations. Starting with the lowest density, CsCl stock solutions containing 2 mg/mL Brij-35 surfactant were loaded stepwise to the outer wall of the spinning rotor forming a nonlinear gradient in-situ. Following gradient formation, a 200-mL aqueous slurry containing8.0 0.05g of coal and 2.0 0.05 g of Brij-35 surfactant was routed to the center of the spinning rotor. After sample loading, the rotor speed was slowly increased to provide a centrifugal force of from 7000-25OOOg across the rotor and maintained for 1 h. Using an immiscible, high-density fluorocarbon, the gradient was then forced from the spinning rotor through a portable density meter used to determine collection cut points. All fractions were vacuum filtered through 0.45 pm pore size filters, thoroughly rinsed with deionized water, dried overnight a t 60 "C under 70-80 kPa vacuum, and weighed. For sample 5445, a total of 40.0 g of coal was processed with a recovery of 38.56 g (96.4%);for 5447, a total of 24.05 g of coal was processed with a recovery of 23.00 g (95.6%); and for 5448, a total of 24.01 g was processed with 23.26 g recovered (96.9%). The rotor effluent from each run was divided into 20 density fractions plus the material that passed through the gradient and "pelleted" against the outer wall of the rotor. The appropriate fractions from repeat DGC runs were composited providing a total of 21 fractions for subsequent analysis for each study sample (Table 4). Petrography of Density Fractions. Splits of each DGC fraction were mixed with a small amount of epoxy and formed into a small plug within a 2.54-cm phenolic ring form. After the epoxy had cured, the ring form was filed with epoxy and allowed

*

*

Ni 218 560 1030 1270 2300 510 Pb 305 305 275 274 428 243

Rb 206 101 38 37 109 175

Sr 630 640 13800 14000 8130 5720

V 620 520 670 900 6040 393

Zn 421 720 298 286 378 422

Zr 226 570 1180 1300 4830 540 ~~

% c1 (wc)

Se (wc)

Br (wc)

0.207

1.39

26.4

0.22 0.289

0.72 0.84

29.45 0.72

to cure. Polishing techniques were modified from those used for the whole-coal pellets, with a coarse grind at 600 grit, followed by final polishes with 0.3- and 0.05-pm alumina slurries. In some cases, hand polishing was necessary due to the small amount of material in the plug. Petrographic analysis was performed on each pellet, counting up to 500 poinb per sample. In some cases (particularly in the lower density fractions) only 200 or 250 points could be counted. P I X E Ash Geochemistry. The major and minor element geochemistry of the high-temperature ashes of all of the lithotypes, the whole coal samples of the three lithotypes subjected to DGC separation, and the DGC density fractions was determined using proton induced X-ray emission (PIXE). Procedures for PIXE analysis can be found in Savage.ll X-ray fluorescence analysis of the lithotype high-temperature ashes was conducted a t the CAER using techniques described by Hower and Bland.5 FTIR Characterization. Fourier-transform infrared (FTIR) spectroscopy was conducted a t the CAER on a Nicolet Model 2OSXC. Procedures for the analysis of the whole-coal fractions were adapted from techniques described by Kuehn.12J3 The organic band widths selected for area integration were as follows: aromatic C-H, 3090-2990 cm-1; aliphatic C-H, 29902740 cm-1; carbonyl C=O, 1770-1680 cm-'; aromatic C=C, 16801520 cm-1; aliphatic C-C, 1520-1385 cm-1; terminal CHs, 13851357 cm-'. Analysis of the low-temperature ashes was adapted from (11)Savage, J. M.; Wong, A. S.; Robertaon, J. D. Thick-targetP E E / PIGE analysis of complex matrices. J . Trace Microprobe Tech. 1992,10, 151-168. (12) Kuehn, D. W.; Davis, A.; Painter, P. C. Characterization of the organic structure of the Lower Kittanning coal seam using Fourier Transform Infrared Spectroscopyand optical properties. Coal Research Section,ThePennsylvania State University, ReportDOE-30013-F4,1988;

241 p. (13)Kuehn, D. W.; Davis, A. The effects of coalification and paleoenvironment on aliphatic and aromatic CH contents in the Lower Kittanning coal seam. Org.Geochem. 1991,17, 255-262.

Lithotypes from the Blue Gem Coal Bed Table 3. Maceral Content of Bench Samples sample vitrinite fusinite semifusinite micrinite exinite resinite 2.3 3.6 0.2 3.7 2.9 5443 87.3 7.7 0.7 2.3 1.8 85.1 2.4 5444 2.8 2.5 10.1 0.1 78.5 6 5445 1.4 2.2 6.7 0.6 85.8 3.3 5446 77.6 7.2 3.6 2.7 8.7 0.2 5447 1.4 2.6 6.2 1.3 83.1 5.4 5448 techniques described by Painter e t al.14 In the procedure used a t the CAER, 1mg of sample was mixed with 300 mg of KBr and ground for 5 min and stored in a desiccator overnight. A 50-mg portion was pressed into a 5-mm pellet. The development of a quantitative mineral reference spectral library required the adoption of an internal standard. The standard was required to have a strong signal in a spectral region free of organic or coal mineral absorption (2700-1800 cm-l) (and weak signals elsewhere), be thermally stable over time, not be reactive with KBr, and not be reactive with the mineral components. Potassium thiocyanate (KCNS) was selected and mixed at 0.20% (weight basis) with KBr. All scanned samples were compared to a reference KCNS/KBr spectrum. Mineral spectra were obtained from an internally prepared library and the commercially available USGS mineral library. A series of quantitative mineral mixes were run in order to standardize the system for the anticipated major minerals: quartz, kaolinite, calcite, siderite, montmorillonite, illite, and gypsum. Once standardized, FTIR spectra were compared with X-ray diffraction results for coal ashes previously analyzed a t the CAER. Spectra were obtained at 2-cm-1 resolution based on 400 scans.

Results Petrographic Analysis of Bench and DGC Samples. Petrographically, the bench samples have relatively high vitrinite contents, with samples containing between 77.6 and 87.3 % vitrinite (Table 3). Coal rank is high-volatile A bituminous (0.88% Rmax). Petrography of the DGC separates is summarized on Figures 1-3. In the DGC fractions recovered between 1.25 and 1.30 g/mL, vitrinite content is over 90%, exceeding 98% in at least one fraction for each series. However, less-dense fractions (95 % vitrinite range. Rimmer et al.I5showed that a variety of vitrinite submacerals exist in this coal (14)Painter, P. C.;Coleman, M. M.; Jenkins, R. G.; Whang, P. W.; Walker, P. L., Jr. Fourier Transform Infrared study of mineral matter in coal. A novel method for quantitative mineralogical analysis. Fuel 1978,57,337-344.

Energy & Fuels, Vol. 8, No. 3, 1994 721

bed which can be identified following etching by techniques described in Stach et a1.7and Stanton and Moore.16 These vitrinite macerals most likely have different origins, different degrees of preservation, and slight differences in chemistry. Previous work on vitrinite submacerals has noted differences in reflectance and chemical composition between vitrinite submacerals, with desmocollinite having a lower reflectance and higher hydrogen content than tel~collinite.~ Criteria for identification are based in part on morphology, a feature which can be elusive in -325mesh particles. Consistent identification of vitrinite types, while desirable, was not deemed to be achievable in this size range. Qualitative observations of the larger particles point toward a shift from a desmocollinite associated with mixed maceral assemblages in the lighter fractions to telocollinite in the denser fractions. With the abrupt transition from vitrinite to semifusinite domination in the density fractions, there is a lesser tendency toward mixed maceral assemblages at higher densities. Geochemical Characterization of Bench Samples. The bench samples show trends typical for the Blue Gem seam in Knox County, Kentucky. Aside from the top bench of the coal bed, KCER-5443, which has an ash content of 9.15% and a total sulfur content of 2.32%, the coal bed is very low in ash (0.40-1.47 % ) and sulfur (less than 0.8 5% ). Previously, we have suggested, based on the extremely low ash and sulfur contents, that the Blue Gem swamp was primarily rainfed, or “raised”, receiving little detrital input throughout most of its d e ~ e l o p m e n t . ~ J ~ Elemental analyses suggest the presence of clays (Si and Al), carbonates (Fe and Ca), and, in the upper bench, possibly pyrite (Fe and S). This is supported by previous observations on the mineralogy of this seam.l5 Several trace elements are enriched in the center of the seam (Ba, Cu, Mo, and Sr); other elements (Cr, Co, Mn, V, Ti, and Zr) increase toward the base of the seam. Rb, which decreases in the middle of the seam, follows the trend for K. These observations are in general agreement with those made by Hower et ala3 Geochemistry of DGC Samples. There is no Cs in the whole-coal and lithotype (non-DGC) ashes, and therefore Cs in the DGC separates appears to be solely from the CsCl salt solutions used to prepare the DGC gradient. The PIXE geochemistry on the whole-coalbasis was calculated on a Cs-free basis. The C1 content on the whole-coal basis was also adjusted stochiometrically for Cs remaining from the DGC processing. The PIXE results for ashed and whole-coal samples are given in Table 2 and for the DGC splits in Table 5. Whole-Coal Basis. C1 and Br, both of which are best evaluated on the whole-coal basis, both maximize in the vitrinite-rich separates (Figure 4). C1 is known to be relatively high in the Blue Gem coal bed with an average of 0.24% in samples from the Barbourville q~adrang1e.l~ C1was not detected by PIXE analysis of the ashed samples. In very-low ash samples such as these, organic sulfur is the most prevalent element other than C, H, N, and 0. Even on the whole-coal basis (pre-DGC),the sulfur content of KCER-5445 exceeds the ash yield. The concentration of organic relative to inorganic matter matter in the DGC (15)Rimmer, S.M.; Moore,T. A.; Esterle,J. S.; Hower, J. C. Geological controls on sulfur content of the Blue Gem coal seam, southeastern Kentucky. Appalachian Basin Industrial Associates, Ninth Meeting, October 17-18,1985,Morgantown, WV, Vol. 9,pp 212-225. (16)Stanton, R. W.; Moore, T. A. Types of vitrinite macerals I; The Org. Pet. Newsletter 1991,8(1),8-10, necessity for etching. SOC. (17)Hower, J.C.; Riley, J. T.; Thomas, G. A.; Griswold,T. B. Chlorine in Kentucky coals. J. Coal Qual. 1991,10, 152.

722 Energy & Fuels, Vol. 8, No. 3, 1994

Hower et al.

Table 4. Ultimate and Proximate Analysis of DGC Separates cut point Coal 5445 44-4-1 44-4-2 44-4-3 44-4-4 44-4-5 44-4-6 44-4-7 44-4-8 44-4-9 44-4-10 44-4-11 44-4-12 44-4-13 44-4-14 44-4-15 44-4-16 44-4-17 44-4-18 44-4-19 44-4-20 44-4-21 Coal 5447 44-4-1 44-4-2 44-4-3 44-4-4 44-1-5 44-1-6 44-1-7 44-1-8 44-1-9 44-1-10 44-1-11 44-1-12 44-1-13 44-4-14 44-1-15 44-1-16 44-1-17 44-1-18 44-1-19 44-1-20 44-1-21 Coal 5448 44-2-1 44-2-2 44-2-3 44-2-4 44-2-5 44-2-6 44-2-7 44-2-8 44-2-9 44-2-10 44-2-11 44-2-12 44-2-13 44-2-14 44-2-15 44-2-16 44-2-17 44-2-18 44-2-19 44-2-20 44-2-21

1.12

1.15 1.18 1.20 1.22 1.24 1.25 1.26 1.27 1.28 1.29 1.30 1.31 1.32 1.34 1.36 1.38 1.40 1.42 1.42+ pellet 1.12

1.15 1.18 1.20 1.22 1.24 1.25 1.26 1.27 1.28 1.29 1.30 1.31 1.32 1.34 1.36 1.38 1.40 1.42 1.42+ pellet 1.12 1.15 1.18 1.20 1.22 1.24 1.25 1.26 1.27 1.28 1.29 1.30 1.31 1.32 1.34 1.36 1.38 1.40 1.42 1.42+ pellet

average density

total weight

(wt%)

total C(%)

H(%) N(%)

1.110 1.140 1.170 1.192 1.215 1.234 1.247 1.257 1.266 1.276 1.286 1.294 1.301 1.311 1.324 1.348 1.368 1.388 1.407 1.440

0.067 0.086 0.287 0.305 1.008 2.393 1.966 2.789 3.496 4.552 6.105 7.697 4.897 0.691 0.305 0.372 0.276 0.211 0.189 0.402 0.468

0.17 0.22 0.74 0.79 2.61 6.20 5.10 7.23 9.07 11.80 15.83 19.96 12.70 1.79 0.79 0.96 0.72 0.55 0.49 1.04 1.21

79.79 83.74 89.55 87.69 86.95 82.01 83.62 82.96 83.02 81.99 81.96 82.17 83.08 89.09 89.97 91.19 91.98 94.34 94.62 92.89 82.19

10.98 7.87 6.89 6.49 6.30 6.06 5.73 5.65 5.37 5.24 5.16 5.00 5.00 4.62 4.36 3.91 3.60 3.35 3.08 3.02 2.39

1.03 0.96 0.77

1.110 1.142 1.176 1.194 1.218 1.238 1.247 1.256 1.268 1.277 1.284 1.291 1.301 1.313 1.327 1.347 1.368 1.388 1.408 1.440

0.022 0.034 0.099 0.138 0.383 1.107 1.132 1.956 2.902 4.227 5.679 2.918 0.328 0.209 0.315 0.213 0.169 0.138 0.118 0.261 0.651

0.10 0.15 0.43 0.60 1.67 4.81 4.92 8.50 12.62 18.38 24.69 12.69 1.43 0.91 1.37 0.93 0.73 0.60 0.51 1.13 2.83

84.86 88.83 86.06 84.98 83.62 82.84 82.84 81.66 82.20 79.13 81.61 76.13 81.61 81.97 70.18 81.38 70.11 84.73 72.33 62.80

6.99 7.44 6.89 6.53 6.34 6.04 5.72 5.62 5.43 5.15 5.10 5.08 4.76 4.48 4.20 3.87 3.64 3.58 3.18 2.23

3.93 2.63 2.13 1.94 2.02 2.00 2.17 2.07 2.07 1.97 1.89 1.64 1.52 1.43 1.15 1.22 1.17 1.43 0.92 0.72

1.110 1.140 1.170 1.192 1.215 1.234 1.247 1.257 1.266 1.276 1.286 1.294 1.301 1.311 1.324 1.348 1.368 1.388 1.407 1.440

0.027 0.041 0.105 0.151 0.403 1.147 1.366 2.581 3.943 4.411 4.018 2.953 0.652 0.162 0.211 0.188 0.107 0.093 0.073 0.155 0.472

0.12 0.18 0.45 0.65 1.73 4.93 5.87 11.10 16.95 18.96 17.28 12.70 2.80 0.70 0.91 0.81 0.46 0.40 0.31 0.67 2.03

86.57 83.29 86.24 83.44 83.50 81.33 81.47 81.53 81.74 81.24 81.67 83.46 84.43 86.07 88.39 88.59 87.23 87.89 88.01 57.80

7.62 7.13 6.67 6.47 6.20 5.88 5.69 5.64 5.40 5.20 5.09 5.08 4.47 4.32 4.05 3.92 3.81 3.51 3.48 2.40

2.53 1.74 2.01 1.90 2.07 1.99 2.09 2.13 2.13 2.05 1.91 2.01 1.90 1.26 1.23 1.60 1.37 1.75 1.28 0.83

fractions enhances the concentration of organic sulfur, the trend observed in the Blue Gem concentrates. Ash Basis. Ash-basis geochemistry is based on several assumptions: (1)the “ash” percentage is 100% minus the “carbon” reported in the PIXE analysis (“carbon”includes all elements lighter than Na); (2) most of the sulfur is organic and not part of the ash, although sulfur can be fixed as sulfate in the low-temperature ashing process; and (3) C1 and Br are volatilized, perhaps as HCl and HBr,

2.56 2.31 1.71 1.80 1.86 1.78 1.93 1.96 2.00 2.07 2.08 1.98 1.81 1.95 1.56 1.36 1.19

S(%)

0.48 0.46 0.43 0.47 0.49 0.47 0.47 0.46

1.11

0.53 0.50 0.49 0.44 0.43 0.46 0.43 0.43 0.43 0.44 0.36 0.02 0.04 0.03 0.03 0.03 0.09 0.21

0.52 0.55 0.50 0.48 0.48 0.49 0.50 0.47 0.47 0.43 0.43 0.47 0.34 0.31 0.27 0.48 1.70

HzO(%)

VM(%) F C ( % )

ash(%)

2.10 0.10 0.10 0.80 0.20 0.30 1.70 1.70 1.90 1.10 1.80 1.50 2.20 2.10 1.80 1.60 1.00 0.60 0.40 0.90 1.20

68.6 59.0 66.1 56.0 51.1 46.0 41.5 41.1 36.7 36.4 35.3 32.6 29.7 31.1 28.2 23.0 20.3 18.6 15.7 16.0 12.9

26.4 39.1 33.5 43.0 48.0 53.2 56.5 56.6 60.7 62.1 62.5 64.9 67.1 66.8 68.7 74.6 77.8 80.5 82.3 82.1 72.4

2.90 1.80 0.40 0.20 0.70 0.50 0.30 0.60 0.65 0.40 0.50 0.20 0.60 0.60 1.20 0.90 0.95 0.80 1.10 1.10 13.60

0.30 0.10 0.90 1.00 1.00 0.50 0.90 1.10 1.00 1.10 2.40 1.70

70.0 66.6 46.1 49.9 46.0 41.5 41.0 39.9 37.5 33.9 31.4 29.5

28.5 32.9 52.5 48.9 52.5 51.2 57.6 57.9 61.0 63.2 65.2 67.6

1.20 0.40 0.50 0.20 0.50 0.80 0.50 1.10 0.50 0.80 0.60 0.60

1.20 0.50 1.00 0.90 0.40 0.50 0.50

26.6 19.0 21.0 21.6 19.1 15.1 11.5

71.6 65.6 68.4 64.5 76.4 65.7 65.3

0.60

0.10 0.10 0.70 0.50 1.00 1.00 0.90 1.30 2.00 0.10 1.50 1.50 2.00 1.10 1.00 1.00 1.00 0.70 0.90 0.80

70.9 64.0 56.8 50.7 45.8 41.8 40.2 39.7 33.0 35.7 24.0 32.5 33.0 28.9 48.2 74.2 89.9 23.2 86.1 57.5

25.7 28.9 37.9 48.2 46.8 56.7 58.3 58.8 64.0 63.8 73.8 64.5 62.9 68.0 48.1 21.3 4.9 71.9 7.8 4.8

3.30 7.00 4.60 0.60 6.40 0.50 0.60 0.20 1.00 0.40 0.70 1.50 2.10 2.00 2.10 3.40 4.20 4.20 5.20 36.70

in the ashing process. Ash-basis geochemical trends are based on the percentages of elements following those corrections. Among the major elements, only P and Mg exhibit lo2 order-of-magnitude ranges in concentration (Figure 5). In the case of Mg, the concentration peaks in the KCER5445 >1.40 g/mL fractions. Ca and Mn are also relatively concentrated through this range, suggesting that the fusinite may be associated with a dolomite or magnesite

Lithotypes from the Blue Gem Coal Bed

Energy & Fuels, Vol. 8, No. 3, 1994 723

pelleb were examined but carbonate phases were not observed (recall that the ash percentage is on the order of 0.1 wt %).

100

Si has a high concentration in the x1.12 g/mL fraction from 5445 (2.9%ash), perhaps representing biogenic silica, and in the pellets (highest density fraction) from the DGC runs. The A1variation is high in 5445 but somewhat muted in 5447 where DGC fraction 1 was not analyzed. K and the K/A1 ratio are low in the vitrinite-rich fractions. K is also higher in 5447 than in 5445. Rb, which can substitute for K, only appears in the high-ash pellets.

20 0

1.3

1.2

1.1

1.5

1.4

Density (g/mL)

I

Liptinite -A- Vitrinite + Inertinite

++

I

Figure 1. Group maceral content of KCER-5445 vs DGC fraction.

-

80-

i

\ I I

i

60A

/

40 -

A

I

Density (g/mL)

+ +

Liptinite

*-Vitrinite

+a-

Inertinite

Figure 2. Group maceral content of KCER-5447 vs DGC fraction.

)c

-'.

80-

f'-

i

\

I w

\

If

20

0

1.I

1.2

1.3

1.5

1.4

Density (g/mL)

I

+ +

analyzed for the other two coals but P was not significantly higher than in the other fractions. We do not know the souce of the P anomaly, but one possibility would be fish scales or bone contributing collophane to the maceralmineral assemblage. On the ash basis, Ti peaks in the high vitrinite fractions with 5447 having higher Ti concentration than 5445. Zr generally follows the Ti trends, peaking in the 5447 1.291.30 g/mL fraction at 6.5% (ash basis).

\f

i

As noted above, P is exceptionally high in the 5445 1.181.22 g/mL fractions. The 1.2Ck1.22 g/mL fraction was

Liptinite -A- Vitrinite

-Q-

Inertinite

1

Figure 3. Group maceral content of KCER-5448 vs DGC fraction.

with Mn substitution. Ca is also high in 5445 and 5447 1.22-1.30 g/mL fractions but drops in the semifusiniterich fractions. Fe is higher in 5445, peaking in the fractions above 1.31 g/mL. Siderite (FeC03), generally dispersed in vitrinite, is an important mineral phase in the Blue Gem.3 The Ca-Fe-Mg trends could be indicative of a shift in the dominant carbonate phase. The mineralogic analysis (below) proved to be inconclusive. The polished

V peaks in the high-vitrinite fractions, with 5447 having much higher concentrations, up to 14% in the 5447 1.291.30 g/mL fraction. In part, Cr followsthe V trend, peaking in the same fraction. Cr also has a high concentration in the 5445 1.12-1.15 g/mL fraction, corresponding to high concentrations of Ni, Cu, Mn, Zn, and Pb. Co and Ni peak on the light side of the high-vitrinite fractions in 5445 while in 5447, where concentrations are higher, Ni continues to be high through the high-vitrinite fractions. In addition to the Cu high in the 5445 1.12-1.15 g/mL fraction, Cu is also high in the 5445 1.18-1.20 and 1.291.30 g/mL fractions. Cu is higher in 5445 than in 5447. In the latter lithotype, it decreases through the highvitrinite range. Zn is particularly high in the 5445