The application of automated mineralogy to the provenance study of red-bed successions: a case study from the Permo-Triassic of SW England

This thesis reviews heavy mineral separation techniques for red-bed heavy mineral provenance studies. It demonstrates the effectiveness of automated mineralogy for mineral comparisons and targeted mineral analysis, resulting in new provenance history conclusions for the Permo-Triassic of SW England. The methodology was developed to remove iron-oxide coatings from grains and provide optimal separation of heavy mineral concentrates in comparison to hydroseparation and panning. Step-wise methodology iterations were validated at each stage using the QEMSCAN (Quantitative Evaluation of Minerals by SCANing electron microscopy). The QEMSCAN works in a similar way to the SEM (scanning electron microscope) and electron-microprobe. Four EDS (energy dispersive spectroscopy) detectors measured X-rays, secondary electrons and back-scattered electrons from the sample. The software allocated these to defined geochemical SIP (species identification protocol) categories. The SIP was back-validated using the electron-microprobe, XRD (X-ray diffraction), microscopy and SEM. QEMSCAN was the primary analysis technique. The results included false colour mineral maps and semi-quantitative statistics for the SIP categories. This enabled targeted mineral analysis with supporting techniques. The methodology was applied to the Permo-Triassic of SW England. The QEMSCAN was able to pick up small scale and large scale heavy mineral provenance trends. Case study 1 showed regional heavy mineral trends and allowed lithological correlation of outliers. Case study 2 used heavy minerals to confirm contemporaneous structural evolution and the base of the Upper Permian in South Devon. Case study 3 allowed a re-evaluation of the base Triassic. Case study 4 targeted biotite minerals at Corbyn’s Head. Case study 5 reviewed local vertical and horizontal trends in comparison to regional trends. This has significant oil and gas exploration implications as it gives confidence in correlating deposits over 10s km and can be applied to North Sea barren red-beds. Finally, some mineral firsts, including rammelsbergite and a topaz rhyolite were identified