Die verband tussen die petrografiese en chemiese eienskappe van Suid-Afrikaanse steenkool

Abstract

The chemical and physical characteristics of coal display a direct relationship to rank and type. Each maceral with a given rank will therefore have a more or less definitive chemical and physical composition. Rank, grade and type can effectively be applied to characterise coal. To a great extent these characteristics determine the technological utilisation potential of coal. European researchers have already identified these characteristics with respect to Northern Hemisphere coal. These correlations cannot, however, be applied directly to South African coal because of the differences in mode of origin of the coal. The objectives of this thesis are to correlate the chemical (carbon, hydrogen, oxygen, volatile matter content and calorific value) and the physical characteristics (HGI) and the coking properties (swelling number and roga index) with each other and with the petrographic characteristics. Coal samples of collieries in the Transvaal, Natal and the Orange Free State have been collected and proximate analyses, ultimate analyses, swelling numbers, HGI, calorific values and petrographic analyses have been carried out on them. Dry ash-free C and H values were originally used to compile a Seyler diagram for South African coal in which the calorific value and volatile matter were related to the elemental composition. Owing to various factors influencing the coal characteristics, the distribution pattern of the volatile matter values is not equidistant as that of the calorific values. Although a linear pattern exists, the isolines do not keep the same distance from one another as is the case with the calorific values with the result that four areas can be distinguished: Area 1 Volatile matter higher than 30%; Area 2 Volatile matter higher than 15% but lower than 30%; Area 3 Volatile matter higher than 6% but lower than 15%; Area 4 Volatile matter lower than 6%. Correlation relationships were found on this Seyler diagram by using dry ash-free volatile matter and calorific values. In the case of certain parameters these relationships were relatively poor. An approach was subsequently followed in which the coalfields were subdivided into geographic areas, namely the Witbank-Middelburg area, the Ermelo-Belfast-Piet Retief area, the Orange Free State area and the Natal area. Correlations have been established with respect to the C, H and O contents, type, rank, tar yield, swelling number, roga index and HGI. Apart from these relationships the connection between Rov( ) and max Rov(rand) was also determined, as well as the correlation between vitrinite and RSF for normal and beneficiated products. These correlation relationships were in general very good and as a result any two determined values (preferably the volatile matter content and calorific value) can be used to derive the values of all the other parameters graphically or it can be calculated with the aid of formulae.