Henry, Delano Gerardo (2019) Assessing Raman spectroscopy to determine the thermal maturity of organic matter : application to hydrocarbon exploration. (PhD thesis), Kingston University, .
Abstract
Determining the thermal alteration (maturity) of organic matter (OM) in sedimentary rocks is vital for predicting hydrocarbon (HC) generation and to calibrate basin models that help petroleum geologists delineate areas of interest for HC exploration. Vitrinite reflectance (VR) is the ‘gold standard’, however uncertainty can arise in certain circumstances, such as in rocks with sparse vitrinite particles, hydrogen-rich OM and from overpressured basins, plus issues relating to sample preparation and operator inexperience. VR is also limited to post-Silurian rocks. It is therefore essential to have a wide range of maturity methods available that an operator can call upon depending on the circumstances. This study demonstrates that Raman spectroscopy is a viable rapid and non-destructive technique to determine OM maturity that requires minimal sample preparation for coals and mudrocks. A new standard acquisition and processing method is described that is rapid and introduces less bias than previous methodologies, as spectral deconvolution is not performed. The method includes performing a Savitzky-Golay smoothing filter using a 21-point quadratic polynomial algorithm, a 3rd-order polynomial baseline correction, followed by normalizing all spectra to a maximum G-band height of 2000 au. An automated spreadsheet is presented that automatically calculates the most commonly used Raman parameters including the G-band full-width at half-maximum (G-FWHM), Raman band separation (RBS), R1 (height ratio of the D1- and G-band) and a new parameter termed the scaled spectrum area (SSA; total area between 1100–1700 cm-1). The method has been tested successfully on a suite of prospective shale gas Carboniferous rocks from the UK with a wide maturity range and has been calibrated with VR. The G-FWHM is the most reliable Raman parameter with a strong correlation with VR (R2 0.96), followed by the SSA (R2 0.88); R1 (R2 0.73) and RBS (R2 0.65). The method has been extended to characterise a suite of shale gas geological reference materials from the United States Geological Survey (USGS) and Jurassic–Cretaceous mudrock samples from the UK and Norwegian continental shelf. It is shown that isolating the organic matter using mineral acids alters the Raman parameters by reducing/removing background fluorescence caused by the non-organic components of the rock. This reduces the bias associated with background subtraction, improving precision and accuracy. Results on samples with known VR retardation due to overpressure demonstrate that Raman may be insensitive to the retardation effect. A portable Raman instrument has also been tested on a set of coal and shale rock-chips samples. The results demonstrate that the portable Raman is more suited to coal samples, as the shale samples have considerable background fluorescence which obscures the OM Raman bands, as well as increasing the bias associated with the background subtraction. A comparison of maturity results from Raman, VR, Rock-Eval (Tmax and production index) and illite crystallinity, illustrate that discrepancies can occur due to technical, methodological and theoretical issues and further supports the consensus that multiple maturity parameters should be used in order reduce the risk in estimating the maturity of source rocks. Raman is a viable alternative method to determine the maturity of OM for HC exploration, but this study highlights the pitfalls and advantages of differing Raman methodologies. It is essential that the analytical community develops a standard approach for the method to be more widely adopted in the future.
![[img]](https://eprints.kingston.ac.uk/style/images/fileicons/other.png)
Actions (Repository Editors)
 |
Item Control Page |