Abstract
Recent developments in the green economy and advancement in the fourth industrial revolution have heightened the need for rare earth elements, expanded to include yttrium and scandium (REY+Sc). The continuous increase in the demand for REY+Sc indicates that there will be an increase in the pressure to supply these elements from conventional and unconventional deposits. A growing body of literature recognises the importance of alternative or unconventional deposits to supplement the supply of these elements, such as extracting REY+Sc from coal and coal byproducts. Hower's research on eastern Kentucky coals revealed a coal deposit with over 2000 ppm REY+Sc (on an ash basis).
Despite the traction of the research conducted on REY+Sc in coal globally, more research needs to be done regarding REY+Sc in South African coal. Work conducted by Wagner, Matiane and Eterigho-Ikelegbe focused on coal ash and coal discards. However, uncertainty still exists about the concentration of REY+Sc in fresh coal and associated sediment samples in the South African context. Hence, this research set out to investigate the REY+Sc concentration and mode of occurrence in coal and associated sediments by comparing three South African coalfields, namely the Ermelo, Witbank and Waterberg coalfields. Furthermore, the study considered whether the REY+Sc concentrations in the three coalfields are potentially economically viable.
Coal and associated sediment samples were obtained from Ermelo (fifteen samples), Witbank (eighteen samples) and Waterberg (twenty samples) coalfields, following borehole core logging. The analytical techniques followed in this research were coal petrography (maceral count and vitrinite reflectance), X-Ray Diffraction (XRD), X-Ray Fluorescence (XRF), sequential chemical extraction procedure (SCEP) and inductively coupled plasma mass spectrometry (ICP-MS), as well as mineral liberation analysis (MLA) on selected samples. The coals from the three coalfields are inertinite-rich, medium-rank bituminous with moderate to high ash content. The two dominant minerals in all samples are kaolinite and quartz.
vi
Among the three coalfield sample sets, the Ermelo samples have the greatest REY+Sc concentration. The ICP-MS results indicated that the Ermelo coal samples had the highest ΣREY+Sc, peaking at 163 μg/g (UNI2M). The Waterberg coal samples yielded the lowest concentration, the lowest being 17μg/g (12A). The associated sediments for the whole sample results, sample UNI2E (Ermelo), have the highest ΣREY+Sc of 644 μg/g. Sample 6A, from the Waterberg, has the lowest ΣREY+Sc, of 26 μg/g. The SCEP results yielded higher REY+Sc compared to the ICP-MS single-step digestion. Moreover, the “carbonates” leaching step yielded the highest REY+Sc for all samples in all three coalfields.
Perhaps the most compelling SCEP finding is from the associated sediment results for the Witbank samples. The “carbonate” and silicates steps yielded high ΣREY+Sc concentrations of 14154.2 μg/g and 14551 μg/g, respectively. The SCEP results provide insight into the impact of various lexiviants used in the overall success of leaching out the REY+Sc. In this research, HCL proved successful in leaching REY+SC from all samples. The lixiviant (HCL) did not leach out the “carbonates” as discussed in literature, as none were detected in the analysed samples. However, “phosphates” were successfully leached out. The MLA results affirm this, as the REY+Sc were hosted in phosphate minerals such as monazite, xenotime and brockite( the latter only detected in the Waterberg sample and not previously reported).
A comparison of the three sample sets results revealed that the REY+Sc concentrations decrease with depth for the Ermelo and Witbank coalfield samples. Contrary to the other coalfields, the Waterberg LREE and MREE values increase with depth, while the Sc and Y concentrations decrease with depth.
The REY+Sc concentration in all the analysed samples proved not economically viable according to the outlook coefficient index (they are unpromising). The results were subsequently mathematically converted to “ash basis” to determine the outlook coefficient index, but the results remained uneconomical, except for the UNI2N Witbank sample. Be that as it may, this study helped create a database on the concentration of REY+Sc in coal and associated sediments across the three coalfields. The concentration may not be economically significant when
analysed on a whole coal basis. However, pre-concentrating the samples via ashing and/or leaching (via SCEP) may increase REY+Sc concentrations by a factor controlled by the ash yield.