Suppression and retardation of vitrinite reflectance, Part 2. Derivation and testing of a kinetic model for suppression

The presence of suppressed and retarded vitrinite reflectance (VR) data int roduces a number of difficulties into the prediction of hydrocarbon generat ion in sedimentary basins. Although the effects of suppression can be remov ed from measured YR values manually, a kinetic model for suppressed vitrini te maturation would enable both suppressed and unsuppressed VB values to be predicted using thermal histories derived from basin modelling. The evalua tion of hydrocarbon generation from suppressed and unsuppressed vitrinite s hows that both have similar reaction kinetics. While hydrocarbon generation involves the rupture of the bonds holding volatiles into the vitrinite str ucture, increases in VR are mainly produced by aromatisation and condensati on reactions which take place after volatiles have been expelled. The react ions involved in hydrocarbon generation are different from those responsibl e for increases in VR, and it is not therefore appropriate to derive kineti c models of vitrinite maturation from laboratory hydrocarbon generation exp eriments. During the maturation of normal (unsuppressed) vitrinite, the volatiles gen erated are expelled via the microporous network; the expulsion efficiency i s not limited by the capacity of the microporous network. In hydrogen-rich (suppressed) vitrinites, excess volatiles saturate the microporous network, restricting further aromatisation and condensation processes within the vi trinite, which results in suppression of VB. Kinetically, this has been mod elled by using a variable pre-exponential or "A" value. Two versions of a k inetic model of vitrinite maturation (SMod-1 and SMod-2) have been prepared , based on the amount of suppression predicted by HI-VR calibration models published by Lo (1993) and Samuelsson and Middleton (1998). Two case studies, involving wells Bunga Orkid-1 (Malay Basin) and 20/3-4 (O uter Moray Firth, North Sea), are discussed. Both wells contain suppressed VR values; well 20/3-4 is also overpressured and contains VR data that are both retarded and suppressed. The application of the SMod model to the well s enables heatflow histories derived from tectonic (rift) histories to be u sed for the prediction of VR data, although in the case of well 20/3-4 the use of a pressure retardation model was also required. Complementary eviden ce to support the use of the heatflow history applied to well 20/3-4 is pro vided by palaeotemperature data obtained from diagenetic concretions.