Physical modelling of chemical compaction, overpressure development, hydraulic fracturing and thrust detachments in organic-rich source rock

摘要

Geological evidence for overpressure is common worldwide, especially in petroleum-rich sedimentary basins. As a result of an increasing emphasis on unconventional resources, new data are becoming available for source rocks. Abnormally high values of pore fluid pressure are especially common within mature source rock, probably as a result of chemical compaction and increases in volume during hydrocarbon generation. To investigate processes of chemical compaction, overpressure development and hydraulic fracturing, we have developed new techniques of physical modelling in a closed system. During the early stages of our work, we built and deformed models in a small rectangular box (40 × 40 × 10 cm), which rested on an electric flatbed heater; but more recently, in order to accommodate large amounts of horizontal shortening, we used a wider box (77 × 75 × 10 cm). Models consisted of horizontal layers of two materials: (1) a mixture of equal initial volumes of silica powder and beeswax micro-spheres, representing source rock, and (2) pure silica powder, representing overburden. By submerging these materials in water, we avoided the high surface tensions, which otherwise develop within pores containing both air and liquids. Also we were able to measure pore fluid pressure in a model well. During heating, the basal temperature of the model surpassed the melting point of beeswax (～62 °C), reaching a maximum of 90 °C. To investigate tectonic contexts of compression or extension, we used a piston to apply horizontal displacements. In experiments where the piston was static, rapid melting led to vertical compaction of the source layer, under the weight of overburden, and to high fluid overpressure (lithostatic or greater). Cross-sections of the models, after cooling, revealed that molten wax had migrated through pore space and into open hydraulic fractures (sills). Most of these sills were horizontal and their roofs bulged upwards, as far as the free surface, presumably in response to internal overpressure and loss of strength of the mixture. We also found that sills were less numerous towards the sides of the box, presumably as a result of boundary effects. In other experiments, in which the piston moved inward, causing compression of the model, sills also formed. However, these were thicker than in static models and some of them were subject to folding or faulting. For experiments, in which we imposed some horizontal shortening, before the wax had started to melt, fore-thrusts and back-thrusts developed across all of the layers near the piston, producing a high-angle prism. In contrast, as soon as the wax melted, overpressure developed within the source layer and a basal detachment appeared beneath it. As a result, thin-skinned thrusts propagated further into the model, producing a low-angle prism. In some experiments, bodies of wax formed imbricate zones within the source layer. Thus, in these experiments, it was the transformation, from solid wax to liquid wax, which led to chemical compaction, overpressure development and hydraulic fracturing, all within a closed system. According to the measurements of overpressure, load transfer was the main mechanism, but volume changes also contributed, producing supra-lithostatic overpressure and therefore tensile failure of the mixture.