Geothermal reservoir characterisation using automated mineralogy; implications for CO2 sequestration

  • Dafydd Evans

    Student thesis: Master's Thesis


    Drill cutting analysis provides a cheaper and less time consuming alternative to the collection of core samples during geothermal exploration, although providing less lithological detail than cores. In this study,drill cuttings collected from the Nesjavellir geothermal field in Iceland were analysed using SEM based automated mineralogy to determine any benefits this methodology may provide for drill cutting analysis in geothermal exploration, such as identification of minerals used in geothermometry, as well as any possible benefits in understanding CO2and H2S sequestration projects. 209 cutting samples were analysed using SEM-EDS and processed using AMICS software at 10m intervals through the NJ-18 well, producing quantitative data for 40 phases, made up of compositionally similar minerals, as well as amorphous material such as volcanic glass. Alongside the quantitative data, mineral maps were produced for each sample, showing the distribution of the different compositional groups which can be directly compared to the analysed thin sections. In addition the data is compared with microscope imagery produced of the cuttings along with XRD data for a selection of samples. Comparison between the samples and the automated mineralogy results appears to show a high degree of accuracy, with textures observed in thin section being well represented in the mineral maps, although some possible overlap is observed between compositionally similar minerals such as zeolites and plagioclase. Using the quantitative automated mineralogy data it is possible to interpret the depths of alteration zones ranging from the smectite-zeolite zone to the epidote-actinolite zone, however, this is hindered by a lack of specific compositional groups for key minerals such as epidote or actinolite, although these may be interpreted from changes in other compositional groups and improved in future studies. A simplified stratigraphic log can also be constructed using the quantitative data for primary phases, comparing samples by abundance of crystalline basalt minerals versus volcanic glass in order to identify lava flows/intrusions, hyaloclastites and mixed lithologies. Accurate quantification of calcite and pyrite also appears to be achieved using this method and may be beneficial for identifying depths where both minerals form readily, which can be used for determining injection depths for CO2and H2S sequestration.
    Date of Award2022
    Original languageEnglish
    SupervisorIan Skilling (Supervisor) & Duncan Pirrie (Supervisor)

    Cite this