Abstract
Fifty-three oil shale samples from the Stuart Deposit in Central Queensland were analysed spectroscopically for hydrocarbon (kerogen) content. Near infrared (NIR) diffuse reflectance spectra of the shale samples exhibited sloping baselines due to particulate scattering and instrumental drift. Multiplicative scatter correction (MSC) and derivative spectroscopy were investigated as means of removing the effects of scattering and non-linear baselines. Partial least squares (PLS) calibration models have been developed, utilising both the entire spectral region as well as narrower chemometrically determined spectral bands. Compared with conventional chemical analysis techniques, NIR diffuse reflectance spectroscopy can provide an efficient, complementary method for the prediction of oil yield from oil shale.
Keywords
Get full access to this article
View all access options for this article.
