South African coals of the Main Karoo Basin, which are generally inertinite-rich, continue to be important to the country’s economic prospects. The formation of inertinite macerals present in coal continues to be a subject of controversial discussion, and is generally attributed to multiple origin pathways. Of these pathways, the inertinite macerals of South African coals have been ascribed to two: 1) aerial oxidation in a cold climate setting; and, 2) charring of plant matter. Although fires were previously considered to have been uncommon in South Africa during the Permian coal-forming period, the presence of fossilized charcoal within the coals of the Witbank Coalfield and the clastic sedimentary rocks of the Vereeniging Coalfield suggests otherwise.
A parent coal sample, comprising of 41.6 vol. % vitrinite and 48.5 vol. % inertinite (mineral-matter-included basis), was obtained from the Witbank Coalfield and density-fractionated to create vitrinite-rich and inertinite-rich sub-samples. The vitrinite-rich sample consists of 81 vol. % vitrinite (dominated by collotelinite and collodetrinite), whereas the inertinite-rich sample consists of 63 vol. % inertinite (dominated by fusinite, semifusinite, and inertodetrinite). Further, the samples comprise of mostly vitrite and inertite microlithotypes. The ESR analysis revealed that the inertinite-rich sample has a higher radical concentration relative to the iso-rank vitrinite-rich sample, implying thermally driven pre-diagenesis metamorphism for the major components of the former sample. According to the NMR analysis, the inertinite-rich sample has an abundance of 6-aromatic carbon rings and the vitrinite-rich sample has multi-ring clusters. The 6-carbons ring in the inertinite-rich sample are interpreted to be guaiacol and syringol, the products of low-temperature (below 400 °C) lignin pyrolysis. Based on the stable isotopes, the vitrinite-rich sample has the lower δ13C and the lower 14N value relative to inertinite-rich counterpart. Because of the higher N-quaternary and N-pyridinic, the loss of 14N in the vitrinite-rich sample is attributed to bacterial degradation. This interpretation is also consistent with the absence of anatomical structure in the dominant vitrinite macerals. In contrast, the inertinite-rich sample has a higher concentration of both N-pyrrolic and N-oxide complexes, with the latter indicative of exposure during charring of the dominant macerals of this sample. It was thus concluded that fusinite and semifusinite present in the No. 4 Seam Upper Witbank coal were formed through charring of plant matter, and that the moisture content of affected vegetation determined the degree of charring, and thus, the resultant inertinite maceral. Inertodetrinite was interpreted to have formed through the same process as the dominant, primary inertinite macerals. Inertodetrinite-forming particles were interpreted to reflect charred matter that was reworked by sedimentary processes. The dominance of monomacerals over bi- and trimacerals was interpreted to reflect an interchange between wet periods during which mostly vitrinite was formed, and relatively dry, inertinite-forming periods during which fires occurred.
According to the py-GC/MS analysis, the vitrinite-rich sample is more chemically diverse than the inertinite-rich counterpart, suggesting that humification and gelification processes result in compounds that give rise to diverse products upon pyrolysis. Nonetheless, the pyrolysis products for the samples generally comprise of similar organic compounds though the inertinite-rich sample has the greater fraction of aromatics and phenolics, also reflected in the 13C CP-MAS SS NMR results. Similarity in pyrolysis products likely reflects similarities in source vegetation for the major macerals present in the samples. Further, similarities in organic molecular chemistry including low and high molecular weight alkanes, alkylbenzenes, alkylphenols, alkyl-subtituted polycyclic aromatic hydrocarbons, and xylene isomers, likely reflect a continuum in chemical composition between vitrinite and the specific inertinite macerals (fusinite, semifusinite, and possibly inertodetrinite). The relative impurity of the density-fractionated samples renders the assignment of pyrolysis products to specific macerals difficult.