A Novel Hydropyrolysis Approach for Optimum Release of Covalently-Bound Biomarker Hydrocarbon Skeletons from Sedimentary Organic Matter

Abstract

In order to generate more representative biomarker profiles, geomacromolecules must first be fragmented into smaller (readily-analysable) structural units to gain access to the covalently-bound pool of biomarkers. Existing chemical degradation and pyrolysis methods for fragmenting and analysing molecular components of kerogen are notoriously poor at providing quantitatively significant and/or accurate data. Chemical degradation methods have been less successfully and less widely applied to kerogen in comparison with solvent-soluble macromolecular fractions (usually applied to polar fractions). In general, since the polar fractions represent only a small proportion of the total OM and the yields of routinely analysable products obtained from chemolysis are low, it follows that the biomarker profiles produced from chemical degradation may also be highly biased with respect to the total biomarker assemblage. With analytical pyrolysis methods, e.g. Py-GC-MS, higher conversions of kerogen can be achieved but the structural information about the molecular constituents that is provided (and hence the accuracy of biogeochemical information that is conveyed) is compromised to a significant extent due to the appreciable secondary reactions (cracking and isomerisation) of released products which manifest at the high pyrolysis temperatures employed (typically >600~ In order to generate more representative biomarker profiles, a far higher proportion of the biomarker constituents must be accessed. Reaction conditions are sought which solubilise a significant proportion of sedimentary OM (approaching 100% conversion), whilst maxiraising the level of structural (biogeochemical) information and minimising the extent of secondary reactions between degradation products. Pyrolysis at high hydrogen pressures (> 10 MPa, hydropyrolysis) eliminates the problem of low yields often associated with the use of sterically-bulky chemical reagents and limits the extent of retrogressive chemistry (leading to char-formation and rearrangement of detectable products) encountered in other analytical pyrolysis methods (e.g. flash-pyrolysis). Fixed-bed hydroyrolysis in the presence of a dispersed sulphided molybdenum catalyst gives rise to overall carbon conversions greater than 85% for petroleum source rocks (Type I and Type II ancient SOM) and high volatile coals (Type III OM), with high selectivities to dichloromethane-soluble tar and low hydrocarbon yields (Roberts et al., 1995). Previous work has demonstrated the unique ability of fixed-bed catalytic hydropyrolysis to release much higher yields of aliphatic biomarker hydrocarbons (including n-hydrocarbons, hopanes, steranes and methyl steranes) from immature kerogens in comparison with solvent extraction, mild catalytic hydrogenation and normal pyrolysis methods (Love et al., 1995, 1996). A combination of slow heating rate (5~ rain1), high hydrogen pressure (15 MPa) and use of a dispersed sulphided molybdenum catalyst represents the best regime for achieving high conversions to dichloromethane-soluble products whilst minimising the structural rearrangement of biomarker species (Love et al., 1997). Staged hydropyrolysis on Goynuk oil shale (NW Turkey, Oligocene, Type I) has confirmed that hopanes released at higher temperatures (above 350~ through cleaving relatively strong bonds (possibly by up to 4 or more ether linkages protected by a macromolecular matrix) are quantitatively more significant than those released at lower temperatures for this immature source rock with a relatively low organic sulphur content (Love et al., 1997). This important pool of (strongly-)bound biomarkers is especially difficult to isolate by existing chemical and thermal degradation methods. The very high yields of products, in combination with the retention of structural information that is observed, dictates that hydropyrolysis will generate a more sensitive and accurate biomarker profile for sedimentary OM than these existing approaches.