We have analyzed elemental and mineralogic contents of 12 peat samples to better understand the factors controlling the nature and distribution of inorganic matter in peat. Peats were collected from a variety of depositional and ecological settings, including marsh environments in Florida and Georgia, raised bogs in Minnesota and Maine, various swamp-forest environments in Florida, Georgia, North Carolina, Minnesota, and New York, and an intertidal environment in southern Florida. Peat type was defined by petrographic analysis. Elemental contents of peats dried at 50 °C were determined using neutron activation; high-temperature ash (HTA) residues were analyzed using x-ray fluorescence. Mineralogies of low-temperature ash (LTA) residues were obtained using x-ray diffraction and scanning electron microscopy. HTA contents of the peat samples ranged from 0.38 to 30.65 wt %. Compositions of HTA residues showed a great range (e.g., Si, 3.5 to 42.5 wt %; Ca, 0.15 to 34.4 wt %; Mg, 0.17 to 20.3 wt %). Mineralogies of LTA residues also varied greatly. Quartz was present in all samples, generally as the dominant crystalline component. Corundum and bassanite, artifacts of the low-temperature ashing procedure, were present in 10 and 8 samples, respectively. Halite was abundant in the sample collected from the intertidal environment. Other minerals present in lesser amounts were alkali feldspar, muscovite, kaolinite, smectite, and pyrite. Poorly crystalline or amorphous material was present in all samples; in some it composed the majority of the LTA residues. The presence of poorly crystalline material made correlation between elemental and mineral contents difficult. Factors controlling the amount and type of inorganic material in these peats are: (1) underlying bedrock or sediments; (2) detrital source areas; and (3) the ecology of the peat-forming botanical communities. Our results show that it is unreliable to generalize peat inorganic contents based solely on geographic province or botanic communities.

The concentration of arsenic (As) in U.S. coal is significant in coal cleaning, coal utilization, and environmental considerations. Arsenic is significant because of its potential toxicity for plants and animals. This chapter examines concentrations and modes of occurrence of arsenic in U.S. coals. The data used in this study are from more than 5,000 determinations of As in coal samples, analyzed on an as-received basis using wet chemical and instrumental neutron activation analysis (INAA). Variation in As content was related to coal provinces and coal regions, coal rank, geologic age, sulfur and ash contents, heating values, and stratigraphic and lateral variation. The As content of foreign coal, roof- and floor-rock and coal partings, and coal wastes is discussed. The mode of occurrence of As is reviewed. It appears that the bulk of arsenic in coal is in sulfide minerals, primarily iron sulfides. Much of the arsenic-bearing sulfides may be epigenetic in origin, or the arsenic may have been emplaced by reaction of pyrite with arsenic-bearing mineralogic solutions. Low levels of arsenic (<5 ppm) may be organically associated. Analysis of the As data indicated the following. (1) Appalachian and Western Interior coals have the highest As content; Fort Union and Wind River regions have the least. (2) There is no systematic relation between As content and coal rank, coal deposit age, and total sulfur content; however, there is a sympathetic relation with pyritic sulfur content. (3) Arsenic concentration is highest in coal with heating values between 12,000 and 13,000 Btu/lb. (4) Stratigraphic profiles of As content for four areas show wide variations, with samples from western Kentucky displaying the least variation and samples from the southern Appalachians the most. Pyritic sulfur-content variations generally parallel those of As content. (5) Arsenic distribution maps of a small area in the Western Kentucky coal field generally show As content increasing from the basin toward the margin. (6) Roof and floor rocks contain similar amounts of As as in U.S. coal; arsenic content in U.S. coal is much less than that in coal wastes. (7) Arsenic enrichment factors reveal little difference on the basis of geologic age and coal rank regardless of whether average shale or the crustal averages are used for comparison. There are significant differences in enrichment factor values between some coal provinces or regions.

Sodium is present primarily as exchangeable cations in lignite, and its concentration is inversely proportional to calcium cations. The presence and concentration of the calcium cations in lignite appear to be related to the calcium ion concentration and permeability of the immediate roof rock, sandstones or glacial drift. Data obtained from several hundred bore holes and field checks in strip mines indicate that there is a good correlation between low sodium content in lignite and the presence of sandy roof rocks.

The Danville Coal Member (VII) of the Dugger Formation marks the top of the Carbondale Group in Indiana. Mineral matter in the Danville Coal is composed mainly of quartz, pyrite, calcite, kaolinite, and illite along with minor amounts of gypsum, anhydrite, smectite, and chlorite. The maceral composition is composed predominantly of vitrinite along with minor quantities of exinite, semifusinite, fusinite, and macrinite and/or micrinite. Trace elements that are enriched in the Danville Coal with respect to their “Clarke” values are Ba, Mg, and Zn. Trace-element analyses indicate that the Danville Coal member is younger than the Upper and Lower Millersburg coals in southern Indiana and suggests that the Hymera Coal Member (VI) was deposited at approximately the same time interval as the Millersburg Coals. The Danville Coal Member was deposited in a forested peat swamp under brackish-water influence with reducing and anaerobic conditions and minimal water cover.

Bench samples from the Springfield (western Kentucky No. 9) coal bed have been analyzed for 25 major, minor, and trace elements, including Au and Pt, by instrumental and radiochemical neutron activation analyses. The elemental enrichment trend of this coal bed, compared with crustal abundances, is similar to the Pittsburgh coal bed. Except for Mn, all the other elements in the western Kentucky No. 9 coal have a larger or comparable range of variation when compared with the Pittsburgh bed. The coefficients of variation increase with decreasing concentrations, or remain constant and independent of concentrations. They could be used in evaluating the mode of occurrence of chemical elements. The abundance of the chemical elements, particularly those at trace concentrations, are believed to have several sources. There are quantities inherent from plant debris, those derived from coal-forming processes, and quantities added from or lost to enclosing rocks during ground-water processes. The quantities from each source are not easily determined, but their source can be approached through the use of elemental ratios of coherent pairs. Analytical data are compared to values interpolated from U.S. Geological Survey isochem maps. Among the eight elements compared, the content of Al, Cr, Na, and Sb appear to correlate well. Three kinds of elemental abundance variation can be recognized. These are regional, local, and in-bed variations. On the basis of study of elemental distributions within mine locations, five variation types are identified. Only types I and V can be contoured directly with simple channel sampling. Local variation of the elements should be assessed for other variation types and composite channel samples should be collected to draw meaningful regional distribution contours. For the purpose of bed identification, coherent triads are presented. The triad Al-La-Sc is the best in characterization of a bed. The triads Br-Cs-Na and Co-Ga-Th are not useful for bed fingerprinting, but they are probably useful to identify coal beds locally.

A new method to determine organic and inorganic associations of trace elements in coal was devised using the ratio of concentrations for 34 elements in five vitrinite concentrates to the concentrations in companion whole-coal samples. Trace-element concentrations were determined by instrumental neutron activation analysis. The samples, all high volatile A (hvA) bituminous coal, were from five different coal beds: two from the Appalachian basin, West Virginia and Alabama, USA; two from the Pennine coalfield, England; and one from the Sydney Basin, Australia. The ash content (weight percent) of the vitrinite concentrates after combustion at 960 to 1,000 °C ranged from 0.6 to 1.3 percent, as compared to 2.5 to 7 percent in the whole coals. Most of the trace elements are enriched in the whole coal relative to the vitrinite concentrates, with some notable exceptions: (1) in the Australian coal, 12 of the 22 elements detected in both the whole-coal and the vitrinite concentrates are enriched in the vitrinite concentrates; (2) arsenic is enriched by 1.2 to 2.2 times in the concentrates in four of five coals; in the remaining coal, from West Virginia, arsenic is enriched in the whole coal by a factor of 22, and the iron is also enriched by a similar factor, which indicate that arsenic is associated with pyrite in this sample; (3) antimony is enriched in the vitrinite concentrates of the three foreign coal samples; (4) tungsten is enriched in the vitrinite concentrates of the four samples where tungsten was detected in both the concentrate and the whole coal; (5) the ash- and shale-normalized rare-earth element (REE) patterns suggest that some of the organic matter is enriched in REE.

Ash-fusion characteristics of the Lower Kittanning seam (western Pennsylvania) can be related to environment of deposition. Non-slagging coals (coals with ashes that have ash-fusion temperatures [AFTs] in excess of 2,600°F) are associated with freshwater environments that occur toward the margins of the basin. Slagging coals (coals with ashes that melt at temperatures less than 2,200°F) occur in the central part of the basin, in areas overlain by shales that have been interpreted to have formed in a brackish environment. Trend-surface analysis indicates that whereas strong basinal trends do exist, locally variability can modify regional trends. High ash-fusion coals are associated with high clay (primarily kaolinite) contents, whereas low-fusion coals are associated with high pyrite and marcasite (and to a lesser extent, siderite) contents. Bivariate analysis of these data shows highly significant negative correlations between AFT and Fe 2 O 3 , pyrite, and siderite. Positive correlations exist between AFT and SiO 2 , Al 2 O 3 , TiO 2 , MgO, and K 2 O. Illite and kaolinite also correlate positively with AFT. An understanding of the oxide and mineral composition of the ash and the depositional environment of the peat can therefore be useful in the prediction of ash-fusion characteristics.

Channel and column samples of the Herrin and Springfield coals, as well as roof shales, were collected from four underground coal mines in southern Illinois. Coal lithotypes (vitrain, bright banded coal, subbright banded coal, dull coal, and fusain) and mineral partings were hand-picked from the channel and column samples. Chemical, petrographic, and mineralogical studies of the samples indicate that the vitrain separates have the highest chlorine (CI; mean 0.52%) and lowest sodium (Na; mean 0.088%) and ash (mean 2.8% high-temperature ash) contents. Roof shale and mineral partings have the lowest CI content (mean 0.25%) and highest Na (mean 0.263%) and ash (mean 40.6% high-temperature ash) contents. The Cl/Na ratio of the lithotypes is significantly greater than that of NaCl, the dominant dissolved salt in the ground water associated with Illinois coals. Chlorine and Na in the coal samples studied are concentrated in vitrinite and mineral matter, respectively. Part of the Na and a small fraction of the Cl occur as dissolved NaCl in macropores and mesopores, mainly within the mineral matter and inertinite. The remaining Na is adsorbed on clay minerals. The zeta potential data obtained for the same lithotypes support the interpretation that the organically associated Cl in coal is adsorbed primarily on the micropore walls. One set of lithotypes was leached with water at 27 °C and 93 °C, and with a 4.3 M NH 4 OH solution at 27 °C. The aqueous leachability of Cl increases from vitrain, through bright banded, subbright banded, and dull coal, to fusain. This is in the order of decreasing vitrinite content and, consequently, of decreasing micropore abundance in which most of the Cl is held. Thus, variable Cl leachabilities among the lithotypes may be related to their pore-size distributions.

Simulated weathering experiments on coals and shales demonstrate that the critical factors responsible for the generation of acid mine drainage (AMD) are the amounts of total sulfur, total carbonate, and the surface area of the pyrite. Total sulfur and carbonate carbon contents differ markedly among paleoenvironments whose distribution has been mapped for the Alleghenian strata of western Pennsylvania. Freshwater ( Estheria- bearing) shales have a mean total sulfur content of 0.15 percent and a mean carbonate carbon content of 0.54 percent. Brackish ( Lingula- bearing) shales have a mean total sulfur content of 2.40 percent and a mean carbonate carbon content of 0.14 percent. Marine ( Chonetes- bearing) shales have a mean total sulfur content of 0.95 percent and a mean carbonate carbon content of 0.63 percent In the simulated weathering experiments, the amount of acidity, sulfate, and total iron exhibit a well-defined positive linear relation with total sulfur in samples whose carbonate carbon content is ≦0.01 percent. Where carbonate carbon contents are >0.01 percent, the amount of acidity, sulfate, and total iron is considerably less, and the linear relation no longer exists. Anomalously high amounts of acidity, sulfate, and total iron were encountered in both samples devoid of and containing carbonate and were associated with samples containing a high relative percentage of framboidal pyrite and/or pyrite having a high specific surface area. Because determination of the percentage of framboidal pyrite is subjective, direct measurement of pyrite surface area is preferred.