USGS Circular 1272: Coal Quality of Selected Korean Coal Basins

USGS: Science for a Changing World - USGS visual identity mark and link to main Web site at http://www.usgs.gov/

Coal Quality of Selected Korean Coal Basins
Poster

By John R. SanFilipo,¹ Alex W. Karlsen,¹ Paul C. Hackley,¹ and Suk Whan Park²

¹U.S. Geological Survey, 956 National Center, Reston, VA 20192 - jsan@usgs.gov
²Korean Institute of Geoscience and Mineral Resources, 30, Gajeong-dong, Yuseong-gu, Daejeon, Korea, 305-350

INTRODUCTION

The U.S. Geological Survey (USGS), in cooperation with the Korean Institute of Geology, Mining, and Materials (KIGAM), is investigating the quality of coals mined in the Republic of Korea. The resulting data will be included in the USGS World Coal Quality Inventory (WoCQI). The primary focus of the study is to establish baseline quality assurance standards between the coal analytical facilities of each agency and to investigate coal quality considerations of Korean coals, notably trace element occurrences. Initial indications are that some Korean coals contain significant quantities of mercury, a potentially Hazardous Air Pollutant (HAP), as well as anomalously high concentrations of rare earth elements (REE) that may be economically recoverable. It is anticipated that this study will lead to additional cooperative investigations between USGS and KIGAM in the areas of REE prospectivity and commercial coal bed methane potential. A coal occurrence map of Korea (fig. 1) and other GIS-based products also are in preparation.

Figure 1. Preliminary map of coal distribution in Korea. Sample locations approximate.

[VIEW / PRINT a high-quality PDF file]

PRELIMINARY RESULTS

Standard Analysis

Eleven samples of run-of-mine coal provided by KIGAM have been received and analyzed by the USGS (table 1). These samples represent all current major coal-mining operations in the Republic of Korea. The mines are all underground operations in structurally complex areas. Of the approximately 4 million tons produced in 1999, about 65 percent was used for electrical generation. The remainder was used commercially and domestically, with a small amount of coke produced from coals of the Tae Meag mine. The coal is high rank, low sulfur, and high ash (commonly in partings).

The coalfields sampled are of Permo-Carboniferous age. Using classical American (ASTM) and European (DIN) parameters, the coals rank from anthracite to meta-anthracite (table 2). These data are consistent with published petrographic data, in particular very high Romin values averaging approximately 4 percent (Park, 1990). In addition to Paleozoic coals, there are high-rank coals of Jurassic age and low-rank Tertiary coals in Korea. None of these are being produced today and have not as yet been sampled for WoCQI.

Elemental Analysis

The 11 samples were analyzed for major elements by Inductively Coupled Plasma-Atomic Emission Spectroscopy (ICP-AES) methods. The results are presented (table 3) as a percentage of the ash, converted to oxides. In order to minimize loss of volatile elements, normal USGS procedures involve ashing the major/trace element split at 525°C, which is lower than the ASTM method of 750°C used for proximate analysis. These samples apparently did not completely combust at 525°C, however, resulting in total oxide determinations of well under 100 percent (51-90 pct). They were therefore re-ashed at 750°C and re-analyzed. The incomplete combustion at the lower temperature was presumably caused by a component of incipient graphitization in these high-rank coals. Note that the ash values reported in table 3 are on an air-dried basis, and are thus higher than those determined for proximate analysis as shown in table 1.

Although the sample size is too small for more than generalized observations, a cursory examination of table 3 indicates that, with the exception of relatively high concentrations of potassium, the distribution of major elements within this sample set is typical for coal ash. The concentrations of the major contributors to potential boiler fouling -- sodium, calcium, and magnesium -- are relatively low for most of these samples. All of these elements are poorly correlated to ash in these samples, and, with the exception of potassium, the correlations are negative (fig. 2), indicating that they are possibly organically bound. Major elements tend to be more reactive when in nonsilicate forms and thus more prone to fouling if they are in sufficient concentration, which would not appear to be the case for this sample set. The Si/Al ratio, a measure of abrasiveness, is also low.

Graph showing elected major elements as a function of ash

Figure 2. Selected major elements as a function of ash. Best fit to magnesium shown as an example of the unusual negative correlation of most major elements to ash in this sample set.

[VIEW / PRINT a high-quality PDF file]

The 11 samples were analyzed for trace and minor elements by ICP-AES or ICP mass spectroscopy (ICP-MS). The results of analysis to date (table 4) indicate that anomalous concentrations exist for only a few trace and minor elements in this data set. Exceptionally and moderately high values are noted in table 4, which takes into consideration the average and maximum U.S. concentrations from Finkelman (1994) and the potential importance of the element for health or economic reasons. The most significant anomaly is the relatively high concentration of mercury in six of the seven samples from the Samcheog coalfield. These six samples average 0.53 ppm, which is more than three times the U.S. average (fig. 3), but are well under the upper recorded limits of 10 ppm in the U.S. and much higher concentrations recorded elsewhere. Because Samcheog accounts for approximately 85 percent of South Korea's coal production, these results are particularly worthy of additional investigation. In this sample set, mercury does not show the usual association with chalcophile elements (figs. 4, 5), and the chalcophiles do not show strong sulfur or iron affinities (figs. 6-8). Other typical sulfide associations in coal ash also do not appear to be present for these coals (fig. 9). Thallium, which commonly occurs in pyrite or with lead-zinc sulfides and is sometimes associated with mercury (Robert Finkelman, personal communication, September 2001), is in consistently low concentrations and further indicates that mercury associations in this data set are atypical (fig. 10).

Figure 3. Histogram of mercury concentrations for conterminous U.S. coal (Tewalt, Bragg, and Finkleman, 2001). Samcheog samples are shown in red along the x axis for comparison. Three of seven samples from the Samcheog field are within the top 5 percent of U.S. mercury concentrations in coal.

[VIEW / PRINT a high-quality PDF file]

Figure 4. Mercury, a potentially hazardous air pollutant (HAP), is commonly associated with chalcophile elements -- especially the potential HAPs arsenic, lead, and selenium-when it occurs in coal. This very small sample set has an atypical slightly negative correlation with these elements, except lead, which is slightly positive, and more so without one extreme value (r =0.4). Pb and Se concentrations are slightly high, the latter generally highest outside of the Samcheog field. Arsenic concentrations are low.

[VIEW / PRINT a high-quality PDF file]

Correlation diagram of sulfur and iron to mercury.

Figure 5. Mercury has a slightly positive correlation with sulfur, but shows no relationship with iron, indicating that it is probably not in solid solution with pyrite. Isolating the Samcheog or Hg-rich samples does not significantly change this relationship. These samples are unusually low in arsenic, providing additional evidence that mercury is not occurring in solid solution with pyrite, with which they are commonly associated as a probable result of hydrothermal alteration.

[VIEW / PRINT a high-quality PDF file]

Correlation diagram of chalcophile elements to sulfur

Figure 6. In addition to having atypical relationships with mercury, the usual chalcophile elements do not appear to behave as such in this sample set . Eliminating the high sulfur extreme (KIGAM-1) does not change these relationships significantly (see inset, same symbols). Much of the iron appears to be in nonsulfide form from a mass balance standpoint as well.

[VIEW / PRINT a high-quality PDF file]

Correlation diagram of other chalcophile elements to sulfur

Figure 7. Other commonly chalcophile elements show similar atypical non-associations with sulfur in this sample set. Eliminating the high sulfur (KIGAM-1) does not change these relationships significantly.

[VIEW / PRINT a high-quality PDF file]

Correlation diagram chalcophile elements to iron

Figure 8. Arsenic, lead, and selenium show weakly negative correlations with iron, also indicating nonassociation with pyrite. Eliminating the high Fe outliers does not significantly alter this relationship.

[VIEW / PRINT a high-quality PDF file]

Correlation diagram of cadmium and gallium to zinc

Figure 9. Cadmium, which is in lower than average concentrations for this data set, does show a strong relationship with zinc, where it frequently occurs as a solid solution in zinc sulfides. Gallium, which is in high concentration in this sample set, shows an atypical low correlation with zinc. Such correlations are normally stronger in enriched concentrations.

[VIEW / PRINT a high-quality PDF file]

Correlation diagram of thallium to mercury

Figure 10. Thallium, a rare element which occurs in pyrite, also shows an atypical negative correlation with mercury in this sample set.

[VIEW / PRINT a high-quality PDF file]

Other consistently high concentrations worth noting in table 4 include gallium and rubidium, which are ubiquitous in the Earth's crust but generally in low quantities, and thorium, which may be associated with REE. Park (1990) noted elevated concentrations of all REE in four Korean coal mines, with a single sample from a mine located just southeast of our sample site KIGAM-1 approaching commercial grade (Robert Finkelman, written communication, DATE). In addition to thorium, some samples contain elevated levels of other elements possibly associated with REE, notably niobium, scandium, and possibly vanadium and cesium. Yttrium, which is generally associated with the lanthanide REE (and, like scandium, is considered a REE by some), is only in average concentration for these samples, but it and other REE-associated elements are all moderately to strongly correlated (figs. 11, 12). This finding indicates a potentially common source, possibly detrital (fig. 13). Analysis of these samples for lanthanides by ICP-MS methods is in progress at USGS. It is also anticipated that additional Korean samples will be obtained for REE analysis. Several other scarce elements show increasing concentration with increasing ash, which may also indicate detrital rather than solution or plant-derived origins (fig. 14), possibly with heavy mineral input (fig. 15). Other potentially valuable byproducts, such as silver and gold, also occur at somewhat elevated levels in some of these samples.

Correlation diagram of rare-earth associated elements

Figure 11. Scandium and yttrium, elements associated with rare earths deposits, show strong correlations in this data set with thorium, an element closely associated with REE occurrences in monazite. Park (1990) reported lanthanides such as cerium in exceptionally high levels in other South Korean samples. Analysis of our samples for the lanthanides is in progress.

[VIEW / PRINT a high-quality PDF file]

Correlation diagram of other rare-earth associated elements

Figure 12. Rare earth-related elements scandium and yttrium are positively correlated in this data set, as are other trace metals that may have a common source

[VIEW / PRINT a high-quality PDF file]

Correlation diagram of rare-earth related elements and zircon

Figure 13. Positive correlation with the rare-earth related elements and zircon may indicate a detrital silicate versus fluid source.

[VIEW / PRINT a high-quality PDF file]

Correlation diagram of selected trace elements with ash

Figure 14. Several trace elements in this data set correlate well with ash, indicating possible detrital origins. This is common for chromium, and could indicate silicate or clay mineral provenance versus solid solution in pyrite. The high correlation of gallium and rare earth-associated elements with ash may be due to detrital origins. The high correlation of ash with rubidium may be atypical of the latter's usual occurrence as a salt. As with sulfide associations, lead shows no distinct trend with ash.

[VIEW / PRINT a high-quality PDF file]

Correlation diagram of vanadium and nickel with chromium

Figure 15. Vanadium, which is in somewhat elevated concentration in this sample set, appears to be strongly correlated with chromium and may be associated with heavy minerals. Chromium shows a fairly strong correlation with nickel, another possible heavy mineral indicator.

[VIEW / PRINT a high-quality PDF file]

All of the associations presented herein are based on a small sample set that includes some extreme values and are not shown to establish statistical significance. Their primary purpose is to guide further analysis, including investigations of occurrence modes by scanning electron microscopy. The modes of occurrence of these elements may affect their mobility and pollution potential.

SELECTED REFERENCES

Chwae, U.C., and others, 1995, Geological map of Korea: Korean Institute of Geology, Mining, and Materials (KIGAM), scale 1:1,000,000.

Finkelman, Robert B., 1994, Abundance, source, and mode of occurrence of the inorganic constituents in coal, in Kural, Orhan, ed., Coal, resources, properties, utilization, pollution: Istanbul, Istanbul Technical University p. 115-125.

Park, Suk-Whan, 1990, Petrology of coals in Samcheog and Chungnam coalfields: Daejeon, Korean Institute of Geology, Mining, and Materials, 142 p., 12 pl. [in English and Korean].

Tewalt, S.J., Bragg, L.J., and Finkelman, R.B., 2001, Mercury in U.S. Coal -- Abundance, distribution, and modes of occurrence: U.S. Geological Survey Fact Sheet FS-095-01, 2 p. (http://pubs.usgs.gov/fs/fs095-01/).

Return to Table of Contents

Coal Quality of Selected Korean Coal Basins Poster