RELATIONSHIPS BETWEEN DATA FROM ROCK-EVAL PYROLYSIS AND PROXIMATE, ULTIMATE, PETROGRAPHIC, AND PHYSICAL ANALYSES OF 142 DIVERSE UNITED-STATES COAL SAMPLES

Basic research on coal and oil shale led to automated pyrolysis analysis of petroleum source rocks; most widely used is the Rock-Eval equipment. In order to interpret Rock-Eval analyses in relation to traditional coal data, we analyzed 142 commercial coals with diverse rank, age, maceral and sulfur contents, for most regions of the United States. We compared the Rock-Eval data with traditional industrial coal data, including volatile matter, calorific value, hydrogen and oxygen content, free swelling index, and vitrinite reflectance. We found: (1) there is a close relationship between T(max) and vitrinite reflectance in the ranges 420-590-degrees-C T(max) and 0.4-3%R(o)max of most coals. (2) A close relationship between T(max) and volatile matter (%VM) extends through the entire sample range, including low-rank samples with 35-70%VM. a range where %VM is not considered to be a useful rank parameter. (3) TOC of medium- and high-rank coals is seriously under-measured by Rock-Eval; TOC of low-rank coals (less than 0.8%R(o)max) is close to ''dry basis'' carbon from ultimate analysis. (4) The direct relationships between oxygen index (OI) and %O and between hydrogen index (HI) and %H are clear, though only broadly defined. However, there is virtually no band of concentrated data points on the HI versus OI pseudo-Van Krevelen diagram comparable to the ''development line'' on the H/C versus O/C diagram. (5) There are systematic relationships between Rock-Eval and industrial coal parameters such as calorific value and FSI, but much standardization would be needed before Rock-Eval could find a place in the coal industry. Tests with blends of coal and quartz sand and with various loads of coal alone showed that the amount of organic matter in the Rock-Eval load greatly influences results. Total load in the crucible, if largely inert, plays a small role, however. Increasing absolute or relative coal content causes under-evaluation of Rock-Eval TOC and over-rating of hydrogen. Blends of several coals yielded hydrogen and oxygen indexes related proportionally to the properties of the individual coals, but T(max) is not raised by addition of high-rank coal until over 40% is added.